library(renv)
library(tidyverse)
library(broom)
library(labelled)
library(psych)
library(GPArotation)
#library(devtools)
#install_github("cran/multicon") # not on CRAN atm
library(multicon)
library(correlation)
library(careless)
library(corrplot)
library(lavaan)
library(semTools)
library(semPlot)
library(knitr)
library(ggdist)
library(ggforce)
library(cowplot)
library(nortest)
library(lmerTest)
library(effectsize)# source("clean_data_st3.R")
# for data protection reasons, we provide the cleaned data sets here where potentially sensitive variables were dropped (see "clean_data_st3.R" for details)
base::load("data/df_sbsa3.rda")Also need this list of the Big Five traits:
### BFI-2 - Traits
b5t_extraversion = c("_01", "_06", "_11", "_16", "_21", "_26",
"_31", "_36", "_41", "_46", "_51", "_56")
b5t_agreeableness = c("_02", "_07", "_12", "_17", "_22", "_27",
"_32", "_37", "_42", "_47", "_52", "_57")
b5t_conscientiousness = c("_03", "_08", "_13", "_18", "_23", "_28",
"_33", "_38", "_43", "_48", "_53", "_58")
b5t_neuroticism = c("_04", "_09", "_14", "_19", "_24", "_29",
"_34", "_39", "_44", "_49", "_54", "_59")
b5t_openness = c("_05", "_10", "_15", "_20", "_25", "_30",
"_35", "_40", "_45", "_50", "_55", "_60")
### BFI-2 - Facets
# facets of extraversion
b5f_sociability = c("_01", "_16", "_31", "_46")
b5f_assertiveness = c("_06", "_21", "_36", "_51")
b5f_energy = c("_11", "_26", "_41", "_56")
# facets of agreeableness
b5f_compassion = c("_02", "_17", "_32", "_47")
b5f_respectfulness = c("_07", "_22", "_37", "_52")
b5f_trust = c("_12", "_27", "_42", "_57")
# facets of conscientiousness
b5f_organization = c("_03", "_18", "_33", "_48")
b5f_productiveness = c("_08", "_23", "_38", "_53")
b5f_responsibility = c("_13", "_28", "_43", "_58")
# facets of neuroticism
b5f_anxiety = c("_04", "_19", "_34", "_49")
b5f_depression = c("_09", "_24", "_39", "_54")
b5f_volatility = c("_14", "_29", "_44", "_59")
# facets of openness
b5f_curiosity = c("_10", "_25", "_40", "_55")
b5f_aesthetic = c("_05", "_20", "_35", "_50")
b5f_imagination = c("_15", "_30", "_45", "_60")
# add keys list (to indicate reverse scoring) - unfortunately this does not work (for the facets) automatically with psych function
keys_extraversion = c("+", "+", "-", "-", "+", "-", "-", "-", "+", "+", "-", "+")
keys_agreeableness = c("+", "+", "-", "-", "-", "+", "+", "-", "-", "-", "+", "+")
keys_conscientiousness = c("-", "-", "+", "+", "-", "-", "+", "+", "+", "-", "+", "-")
keys_neuroticism = c("-", "-", "+", "+", "-", "-", "+", "+", "-", "-", "+", "+")
keys_openness = c("-", "+", "+", "+", "-", "-", "+", "+", "-", "-", "-", "+")
# facets of extraversion
keys_sociability = c("+", "-", "-", "+")
keys_assertiveness = c("+", "+", "-", "-")
keys_energy = c("-", "-", "+", "+")
# facets of agreeableness
keys_compassion = c("+", "-", "+", "-")
keys_respectfulness = c("+", "-", "-", "+")
keys_trust = c("-", "+", "-", "+")
# facets of conscientiousness
keys_organization = c("-", "+", "+", "-")
keys_productiveness = c("-", "-", "+", "+")
keys_responsibility = c("+", "-", "+", "-")
# facets of neuroticism
keys_anxiety = c("-", "+", "+", "-")
keys_depression = c("-", "-", "+", "+")
keys_volatility = c("+", "-", "-", "+")
# facets of openness
keys_curiosity = c("+", "-", "+", "-")
keys_aesthetic = c("-", "+", "+", "-")
keys_imagination = c("+", "-", "-", "+")
b5_vars <- list(list(b5t_extraversion, keys_extraversion),
list(b5t_agreeableness, keys_agreeableness),
list(b5t_conscientiousness, keys_conscientiousness),
list(b5t_neuroticism, keys_neuroticism),
list(b5t_openness, keys_openness),
list(b5f_sociability, keys_sociability),
list(b5f_assertiveness, keys_assertiveness),
list(b5f_energy, keys_energy),
list(b5f_compassion, keys_compassion),
list(b5f_respectfulness, keys_respectfulness),
list(b5f_trust, keys_trust),
list(b5f_organization, keys_organization),
list(b5f_productiveness, keys_productiveness),
list(b5f_responsibility, keys_responsibility),
list(b5f_anxiety, keys_anxiety),
list(b5f_depression, keys_depression),
list(b5f_volatility, keys_volatility),
list(b5f_curiosity, keys_curiosity),
list(b5f_aesthetic, keys_aesthetic),
list(b5f_imagination, keys_imagination))
names(b5_vars) = c("extraversion", "agreeableness", "conscientiousness", "neuroticism", "openness",
"sociability", "assertiveness", "energy",
"compassion", "respectfulness", "trust",
"organization", "productiveness", "responsibility",
"anxiety", "depression", "volatility",
"curiosity", "aesthetic", "imagination")Reshape data
df_sbsa3_wide_wb <- df_sbsa3 %>%
arrange(pid, time) %>%
select(pid, time, group, starts_with(c("sw06", "ml01", "rs01", "sc01"))) %>%
pivot_wider(names_from = time,
names_sep = "_t",
values_from = starts_with(c("sw06", "ml01", "rs01", "sc01")))
# colnames(df_sbsa3_wide_wb)Check CFA item loadings at T1
cfa_meaning <- '
# Define the latent factors
meaning1 =~ NA*ml01_01_t1 + lambda1*ml01_01_t1 + lambda2*ml01_02_t1 + lambda3*ml01_03_t1 + lambda4*ml01_04_t1 + lambda5*ml01_05_t1 + lambda6*ml01_06_t1 + lambda7*ml01_07_t1 + lambda8*ml01_08_t1 + lambda9*ml01_09_t1 + lambda10*ml01_10_t1
# Intercepts
ml01_01_t1 ~ i1*1
ml01_02_t1 ~ 1
ml01_03_t1 ~ 1
ml01_04_t1 ~ 1
ml01_05_t1 ~ 1
ml01_06_t1 ~ 1
ml01_07_t1 ~ 1
ml01_08_t1 ~ 1
ml01_09_t1 ~ 1
ml01_10_t1 ~ 1
# Unique Variances
ml01_01_t1 ~~ ml01_01_t1
ml01_02_t1 ~~ ml01_02_t1
ml01_03_t1 ~~ ml01_03_t1
ml01_04_t1 ~~ ml01_04_t1
ml01_05_t1 ~~ ml01_05_t1
ml01_06_t1 ~~ ml01_06_t1
ml01_07_t1 ~~ ml01_07_t1
ml01_08_t1 ~~ ml01_08_t1
ml01_09_t1 ~~ ml01_09_t1
ml01_10_t1 ~~ ml01_10_t1
# Latent Variable Means
meaning1 ~ 0*1
# Latent Variable Variances and Covariance
meaning1 ~~ 1*meaning1
'
fit_cfa_meaning <- cfa(cfa_meaning, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_cfa_meaning, fit.measures = TRUE)lavaan 0.6.15 ended normally after 22 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 30
Used Total
Number of observations 529 530
Number of missing patterns 1
Model Test User Model:
Test statistic 1867.317
Degrees of freedom 35
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 3832.801
Degrees of freedom 45
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.516
Tucker-Lewis Index (TLI) 0.378
Robust Comparative Fit Index (CFI) 0.516
Robust Tucker-Lewis Index (TLI) 0.378
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -9267.963
Loglikelihood unrestricted model (H1) -8334.305
Akaike (AIC) 18595.927
Bayesian (BIC) 18724.056
Sample-size adjusted Bayesian (SABIC) 18628.828
Root Mean Square Error of Approximation:
RMSEA 0.315
90 Percent confidence interval - lower 0.303
90 Percent confidence interval - upper 0.327
P-value H_0: RMSEA <= 0.050 0.000
P-value H_0: RMSEA >= 0.080 1.000
Robust RMSEA 0.315
90 Percent confidence interval - lower 0.303
90 Percent confidence interval - upper 0.327
P-value H_0: Robust RMSEA <= 0.050 0.000
P-value H_0: Robust RMSEA >= 0.080 1.000
Standardized Root Mean Square Residual:
SRMR 0.247
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
meaning1 =~
m01_01_ (lmb1) 1.390 0.060 23.205 0.000
m01_02_ (lmb2) -0.419 0.071 -5.910 0.000
m01_03_ (lmb3) -0.273 0.074 -3.684 0.000
m01_04_ (lmb4) 1.447 0.058 24.936 0.000
m01_05_ (lmb5) 1.336 0.060 22.400 0.000
m01_06_ (lmb6) 1.484 0.061 24.365 0.000
m01_07_ (lmb7) -0.234 0.073 -3.223 0.001
m01_08_ (lmb8) -0.457 0.075 -6.120 0.000
m01_09_ (lmb9) -1.408 0.074 -18.912 0.000
m01_10_ (lm10) -0.770 0.077 -9.991 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.ml01_01_1 (i1) 4.386 0.072 60.690 0.000
.ml01_02_1 5.178 0.069 75.523 0.000
.ml01_03_1 5.002 0.071 70.793 0.000
.ml01_04_1 4.087 0.072 56.855 0.000
.ml01_05_1 4.699 0.071 66.177 0.000
.ml01_06_1 4.181 0.075 55.923 0.000
.ml01_07_1 4.888 0.069 70.633 0.000
.ml01_08_1 4.843 0.072 67.143 0.000
.ml01_09_1 3.711 0.084 44.107 0.000
.ml01_10_1 4.677 0.077 60.690 0.000
meaning1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.ml01_01_t1 0.831 0.064 13.005 0.000
.ml01_02_t1 2.311 0.143 16.128 0.000
.ml01_03_t1 2.566 0.158 16.209 0.000
.ml01_04_t1 0.640 0.055 11.600 0.000
.ml01_05_t1 0.884 0.066 13.379 0.000
.ml01_06_t1 0.756 0.063 12.081 0.000
.ml01_07_t1 2.479 0.153 16.222 0.000
.ml01_08_t1 2.544 0.158 16.108 0.000
.ml01_09_t1 1.762 0.120 14.673 0.000
.ml01_10_t1 2.548 0.161 15.840 0.000
meaning1 1.000 tidy(fit_cfa_meaning) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 10 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 meaning1 =~ ml01_06_t1 =~ lambda6 1.48 0.0609 0.863
2 meaning1 =~ ml01_04_t1 =~ lambda4 1.45 0.0580 0.875
3 meaning1 =~ ml01_09_t1 =~ lambda9 -1.41 0.0744 -0.728
4 meaning1 =~ ml01_01_t1 =~ lambda1 1.39 0.0599 0.836
5 meaning1 =~ ml01_05_t1 =~ lambda5 1.34 0.0596 0.818
6 meaning1 =~ ml01_10_t1 =~ lambda10 -0.770 0.0771 -0.434
7 meaning1 =~ ml01_08_t1 =~ lambda8 -0.457 0.0747 -0.275
8 meaning1 =~ ml01_02_t1 =~ lambda2 -0.419 0.0709 -0.266
9 meaning1 =~ ml01_03_t1 =~ lambda3 -0.273 0.0741 -0.168
10 meaning1 =~ ml01_07_t1 =~ lambda7 -0.234 0.0727 -0.147Reverse-code and form parcels:
df_sbsa3_wide_wb <- df_sbsa3_wide_wb %>%
mutate(ml01_09_t1_r = ml01_09_t1,
ml01_10_t1_r = ml01_10_t1,
ml01_02_t1_r = ml01_02_t1,
ml01_08_t1_r = ml01_08_t1,
ml01_07_t1_r = ml01_07_t1,
ml01_03_t1_r = ml01_03_t1,
ml01_09_t2_r = ml01_09_t2,
ml01_10_t2_r = ml01_10_t2,
ml01_02_t2_r = ml01_02_t2,
ml01_08_t2_r = ml01_08_t2,
ml01_07_t2_r = ml01_07_t2,
ml01_03_t2_r = ml01_03_t2) %>%
mutate(across(intersect(starts_with("ml01"), ends_with("_r")),
~ recode(.x, `1` = 7L, `2` = 6L, `3` = 5L, `4` = 4L, `5` = 3L, `6` = 2L, `7` = 1L, .default = NA_integer_))) %>%
mutate(meaning_par1_t1 = rowMeans(across(c(ml01_06_t1, ml01_10_t1_r, ml01_03_t1_r, ml01_07_t1_r)), na.rm=T),
meaning_par2_t1 = rowMeans(across(c(ml01_04_t1, ml01_05_t1, ml01_02_t1_r)), na.rm=T),
meaning_par3_t1 = rowMeans(across(c(ml01_09_t1_r, ml01_01_t1, ml01_08_t1_r)), na.rm=T),
meaning_par1_t2 = rowMeans(across(c(ml01_06_t2, ml01_10_t2_r, ml01_03_t2_r, ml01_07_t2_r)), na.rm=T),
meaning_par2_t2 = rowMeans(across(c(ml01_04_t2, ml01_05_t2, ml01_02_t2_r)), na.rm=T),
meaning_par3_t2 = rowMeans(across(c(ml01_09_t2_r, ml01_01_t2, ml01_08_t2_r)), na.rm=T))Check CFA item loadings at T1
cfa_selfes <- '
# Define the latent factors
selfes1 =~ NA*rs01_01_t1 + lambda1*rs01_01_t1 + lambda2*rs01_02_t1 + lambda3*rs01_03_t1 + lambda4*rs01_04_t1 + lambda5*rs01_05_t1 + lambda6*rs01_06_t1 + lambda7*rs01_07_t1 + lambda8*rs01_08_t1 + lambda9*rs01_09_t1 + lambda10*rs01_10_t1
# Intercepts
rs01_01_t1 ~ i1*1
rs01_02_t1 ~ 1
rs01_03_t1 ~ 1
rs01_04_t1 ~ 1
rs01_05_t1 ~ 1
rs01_06_t1 ~ 1
rs01_07_t1 ~ 1
rs01_08_t1 ~ 1
rs01_09_t1 ~ 1
rs01_10_t1 ~ 1
# Unique Variances
rs01_01_t1 ~~ rs01_01_t1
rs01_02_t1 ~~ rs01_02_t1
rs01_03_t1 ~~ rs01_03_t1
rs01_04_t1 ~~ rs01_04_t1
rs01_05_t1 ~~ rs01_05_t1
rs01_06_t1 ~~ rs01_06_t1
rs01_07_t1 ~~ rs01_07_t1
rs01_08_t1 ~~ rs01_08_t1
rs01_09_t1 ~~ rs01_09_t1
rs01_10_t1 ~~ rs01_10_t1
# Latent Variable Means
selfes1 ~ 0*1
# Latent Variable Variances and Covariance
selfes1 ~~ 1*selfes1
'
fit_cfa_selfes <- cfa(cfa_selfes, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_cfa_selfes, fit.measures = TRUE)lavaan 0.6.15 ended normally after 16 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 30
Used Total
Number of observations 529 530
Number of missing patterns 1
Model Test User Model:
Test statistic 312.036
Degrees of freedom 35
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 2853.989
Degrees of freedom 45
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.901
Tucker-Lewis Index (TLI) 0.873
Robust Comparative Fit Index (CFI) 0.901
Robust Tucker-Lewis Index (TLI) 0.873
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -7113.195
Loglikelihood unrestricted model (H1) -6957.177
Akaike (AIC) 14286.389
Bayesian (BIC) 14414.519
Sample-size adjusted Bayesian (SABIC) 14319.290
Root Mean Square Error of Approximation:
RMSEA 0.122
90 Percent confidence interval - lower 0.110
90 Percent confidence interval - upper 0.135
P-value H_0: RMSEA <= 0.050 0.000
P-value H_0: RMSEA >= 0.080 1.000
Robust RMSEA 0.122
90 Percent confidence interval - lower 0.110
90 Percent confidence interval - upper 0.135
P-value H_0: Robust RMSEA <= 0.050 0.000
P-value H_0: Robust RMSEA >= 0.080 1.000
Standardized Root Mean Square Residual:
SRMR 0.058
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
selfes1 =~
r01_01_ (lmb1) 0.865 0.046 18.773 0.000
r01_02_ (lmb2) -1.004 0.052 -19.321 0.000
r01_03_ (lmb3) 0.518 0.037 14.161 0.000
r01_04_ (lmb4) 0.604 0.041 14.894 0.000
r01_05_ (lmb5) -1.003 0.049 -20.544 0.000
r01_06_ (lmb6) -1.141 0.051 -22.462 0.000
r01_07_ (lmb7) 0.573 0.041 13.926 0.000
r01_08_ (lmb8) -0.709 0.054 -13.125 0.000
r01_09_ (lmb9) -1.159 0.048 -23.923 0.000
r01_10_ (lm10) 0.881 0.043 20.303 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.rs01_01_1 (i1) 3.193 0.052 61.496 0.000
.rs01_02_1 3.178 0.059 53.827 0.000
.rs01_03_1 3.996 0.039 103.400 0.000
.rs01_04_1 3.871 0.043 89.531 0.000
.rs01_05_1 2.711 0.057 47.940 0.000
.rs01_06_1 3.144 0.061 51.941 0.000
.rs01_07_1 3.777 0.043 87.144 0.000
.rs01_08_1 3.554 0.056 62.922 0.000
.rs01_09_1 2.584 0.059 43.701 0.000
.rs01_10_1 3.361 0.050 67.200 0.000
selfes1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.rs01_01_t1 0.678 0.046 14.663 0.000
.rs01_02_t1 0.835 0.058 14.477 0.000
.rs01_03_t1 0.521 0.034 15.473 0.000
.rs01_04_t1 0.625 0.041 15.390 0.000
.rs01_05_t1 0.685 0.048 14.172 0.000
.rs01_06_t1 0.635 0.048 13.220 0.000
.rs01_07_t1 0.666 0.043 15.534 0.000
.rs01_08_t1 1.185 0.076 15.667 0.000
.rs01_09_t1 0.506 0.040 12.566 0.000
.rs01_10_t1 0.548 0.039 14.207 0.000
selfes1 1.000 tidy(fit_cfa_selfes) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 10 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 selfes1 =~ rs01_09_t1 =~ lambda9 -1.16 0.0485 -0.852
2 selfes1 =~ rs01_06_t1 =~ lambda6 -1.14 0.0508 -0.820
3 selfes1 =~ rs01_02_t1 =~ lambda2 -1.00 0.0520 -0.740
4 selfes1 =~ rs01_05_t1 =~ lambda5 -1.00 0.0488 -0.772
5 selfes1 =~ rs01_10_t1 =~ lambda10 0.881 0.0434 0.766
6 selfes1 =~ rs01_01_t1 =~ lambda1 0.865 0.0461 0.724
7 selfes1 =~ rs01_08_t1 =~ lambda8 -0.709 0.0540 -0.546
8 selfes1 =~ rs01_04_t1 =~ lambda4 0.604 0.0405 0.607
9 selfes1 =~ rs01_07_t1 =~ lambda7 0.573 0.0411 0.574
10 selfes1 =~ rs01_03_t1 =~ lambda3 0.518 0.0366 0.583Reverse-code and form parcels:
df_sbsa3_wide_wb <- df_sbsa3_wide_wb %>%
mutate(rs01_09_t1_r = rs01_09_t1,
rs01_06_t1_r = rs01_06_t1,
rs01_02_t1_r = rs01_02_t1,
rs01_05_t1_r = rs01_05_t1,
rs01_08_t1_r = rs01_08_t1,
rs01_09_t2_r = rs01_09_t2,
rs01_06_t2_r = rs01_06_t2,
rs01_02_t2_r = rs01_02_t2,
rs01_05_t2_r = rs01_05_t2,
rs01_08_t2_r = rs01_08_t2) %>%
mutate(across(intersect(starts_with("rs01"), ends_with("_r")),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(selfes_par1_t1 = rowMeans(across(c(rs01_09_t1_r, rs01_01_t1, rs01_07_t1, rs01_03_t1)), na.rm=T),
selfes_par2_t1 = rowMeans(across(c(rs01_06_t1_r, rs01_10_t1, rs01_04_t1)), na.rm=T),
selfes_par3_t1 = rowMeans(across(c(rs01_02_t1_r, rs01_05_t1_r, rs01_08_t1_r)), na.rm=T),
selfes_par1_t2 = rowMeans(across(c(rs01_09_t2_r, rs01_01_t2, rs01_07_t2, rs01_03_t2)), na.rm=T),
selfes_par2_t2 = rowMeans(across(c(rs01_06_t2_r, rs01_10_t2, rs01_04_t2)), na.rm=T),
selfes_par3_t2 = rowMeans(across(c(rs01_02_t2_r, rs01_05_t2_r, rs01_08_t2_r)), na.rm=T))Check CFA item loadings at T1
cfa_concept <- '
# Define the latent factors
concept1 =~ NA*sc01_01_t1 + lambda1*sc01_01_t1 + lambda2*sc01_02_t1 + lambda3*sc01_03_t1 + lambda4*sc01_04_t1 + lambda5*sc01_05_t1 + lambda6*sc01_06_t1 + lambda7*sc01_07_t1 + lambda8*sc01_08_t1 + lambda9*sc01_09_t1 + lambda10*sc01_10_t1 + lambda11*sc01_11_t1 + lambda12*sc01_12_t1
# Intercepts
sc01_01_t1 ~ i1*1
sc01_02_t1 ~ 1
sc01_03_t1 ~ 1
sc01_04_t1 ~ 1
sc01_05_t1 ~ 1
sc01_06_t1 ~ 1
sc01_07_t1 ~ 1
sc01_08_t1 ~ 1
sc01_09_t1 ~ 1
sc01_10_t1 ~ 1
sc01_11_t1 ~ 1
sc01_12_t1 ~ 1
# Unique Variances
sc01_01_t1 ~~ sc01_01_t1
sc01_02_t1 ~~ sc01_02_t1
sc01_03_t1 ~~ sc01_03_t1
sc01_04_t1 ~~ sc01_04_t1
sc01_05_t1 ~~ sc01_05_t1
sc01_06_t1 ~~ sc01_06_t1
sc01_07_t1 ~~ sc01_07_t1
sc01_08_t1 ~~ sc01_08_t1
sc01_09_t1 ~~ sc01_09_t1
sc01_10_t1 ~~ sc01_10_t1
sc01_11_t1 ~~ sc01_11_t1
sc01_12_t1 ~~ sc01_12_t1
# Latent Variable Means
concept1 ~ 0*1
# Latent Variable Variances and Covariance
concept1 ~~ 1*concept1
'
fit_cfa_concept <- cfa(cfa_concept, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_cfa_concept, fit.measures = TRUE)lavaan 0.6.15 ended normally after 15 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 529 530
Number of missing patterns 1
Model Test User Model:
Test statistic 244.803
Degrees of freedom 54
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 3007.929
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.935
Tucker-Lewis Index (TLI) 0.921
Robust Comparative Fit Index (CFI) 0.935
Robust Tucker-Lewis Index (TLI) 0.921
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -9254.579
Loglikelihood unrestricted model (H1) -9132.177
Akaike (AIC) 18581.157
Bayesian (BIC) 18734.913
Sample-size adjusted Bayesian (SABIC) 18620.639
Root Mean Square Error of Approximation:
RMSEA 0.082
90 Percent confidence interval - lower 0.071
90 Percent confidence interval - upper 0.092
P-value H_0: RMSEA <= 0.050 0.000
P-value H_0: RMSEA >= 0.080 0.620
Robust RMSEA 0.082
90 Percent confidence interval - lower 0.071
90 Percent confidence interval - upper 0.092
P-value H_0: Robust RMSEA <= 0.050 0.000
P-value H_0: Robust RMSEA >= 0.080 0.620
Standardized Root Mean Square Residual:
SRMR 0.038
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
concept1 =~
s01_01_ (lmb1) 0.936 0.049 19.105 0.000
s01_02_ (lmb2) 1.030 0.051 20.317 0.000
s01_03_ (lmb3) 1.000 0.052 19.057 0.000
s01_04_ (lmb4) 1.044 0.052 20.007 0.000
s01_05_ (lmb5) 0.844 0.051 16.465 0.000
s01_06_ (lmb6) 0.081 0.055 1.474 0.140
s01_07_ (lmb7) 0.736 0.055 13.311 0.000
s01_08_ (lmb8) 0.976 0.045 21.672 0.000
s01_09_ (lmb9) 0.966 0.049 19.624 0.000
s01_10_ (lm10) 0.848 0.053 16.066 0.000
s01_11_ (lm11) -0.795 0.046 -17.144 0.000
s01_12_ (lm12) 0.868 0.053 16.325 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.sc01_01_1 (i1) 3.102 0.055 55.942 0.000
.sc01_02_1 3.361 0.058 57.564 0.000
.sc01_03_1 3.208 0.059 54.067 0.000
.sc01_04_1 2.987 0.060 49.891 0.000
.sc01_05_1 2.832 0.056 50.670 0.000
.sc01_06_1 2.941 0.053 55.600 0.000
.sc01_07_1 2.656 0.058 45.890 0.000
.sc01_08_1 2.597 0.053 48.970 0.000
.sc01_09_1 2.696 0.056 47.962 0.000
.sc01_10_1 2.845 0.057 49.746 0.000
.sc01_11_1 3.435 0.051 67.250 0.000
.sc01_12_1 2.945 0.058 50.886 0.000
concept1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.sc01_01_t1 0.751 0.052 14.547 0.000
.sc01_02_t1 0.744 0.053 14.130 0.000
.sc01_03_t1 0.863 0.059 14.557 0.000
.sc01_04_t1 0.806 0.056 14.271 0.000
.sc01_05_t1 0.941 0.062 15.179 0.000
.sc01_06_t1 1.474 0.091 16.257 0.000
.sc01_07_t1 1.230 0.079 15.627 0.000
.sc01_08_t1 0.535 0.039 13.661 0.000
.sc01_09_t1 0.737 0.051 14.438 0.000
.sc01_10_t1 1.011 0.067 15.200 0.000
.sc01_11_t1 0.747 0.050 15.037 0.000
.sc01_12_t1 1.019 0.067 15.215 0.000
concept1 1.000 tidy(fit_cfa_concept) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 concept1 =~ sc01_04_t1 =~ lambda4 1.04 0.0522 0.758
2 concept1 =~ sc01_02_t1 =~ lambda2 1.03 0.0507 0.767
3 concept1 =~ sc01_03_t1 =~ lambda3 1.00 0.0524 0.732
4 concept1 =~ sc01_08_t1 =~ lambda8 0.976 0.0451 0.800
5 concept1 =~ sc01_09_t1 =~ lambda9 0.966 0.0492 0.747
6 concept1 =~ sc01_01_t1 =~ lambda1 0.936 0.0490 0.734
7 concept1 =~ sc01_12_t1 =~ lambda12 0.868 0.0531 0.652
8 concept1 =~ sc01_10_t1 =~ lambda10 0.848 0.0528 0.645
9 concept1 =~ sc01_05_t1 =~ lambda5 0.844 0.0512 0.656
10 concept1 =~ sc01_11_t1 =~ lambda11 -0.795 0.0464 -0.677
11 concept1 =~ sc01_07_t1 =~ lambda7 0.736 0.0553 0.553
12 concept1 =~ sc01_06_t1 =~ lambda6 0.0812 0.0551 0.0667Reverse-code and form parcels:
df_sbsa3_wide_wb <- df_sbsa3_wide_wb %>% # based on the item content, it makes more sense here to recode all except sc01_11
mutate(sc01_01_t1_r = sc01_01_t1,
sc01_02_t1_r = sc01_02_t1,
sc01_03_t1_r = sc01_03_t1,
sc01_04_t1_r = sc01_04_t1,
sc01_05_t1_r = sc01_05_t1,
sc01_06_t1_r = sc01_06_t1,
sc01_07_t1_r = sc01_07_t1,
sc01_08_t1_r = sc01_08_t1,
sc01_09_t1_r = sc01_09_t1,
sc01_10_t1_r = sc01_10_t1,
sc01_12_t1_r = sc01_12_t1,
sc01_01_t2_r = sc01_01_t2,
sc01_02_t2_r = sc01_02_t2,
sc01_03_t2_r = sc01_03_t2,
sc01_04_t2_r = sc01_04_t2,
sc01_05_t2_r = sc01_05_t2,
sc01_06_t2_r = sc01_06_t2,
sc01_07_t2_r = sc01_07_t2,
sc01_08_t2_r = sc01_08_t2,
sc01_09_t2_r = sc01_09_t2,
sc01_10_t2_r = sc01_10_t2,
sc01_12_t2_r = sc01_12_t2) %>%
mutate(across(intersect(starts_with("sc01"), ends_with("_r")),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(concept_par1_t1 = rowMeans(across(c(sc01_04_t1_r, sc01_01_t1_r, sc01_05_t1_r, sc01_06_t1_r)), na.rm=T),
concept_par2_t1 = rowMeans(across(c(sc01_02_t1_r, sc01_09_t1_r, sc01_10_t1_r, sc01_07_t1_r)), na.rm=T),
concept_par3_t1 = rowMeans(across(c(sc01_03_t1_r, sc01_08_t1_r, sc01_12_t1_r, sc01_11_t1)), na.rm=T),
concept_par1_t2 = rowMeans(across(c(sc01_04_t2_r, sc01_01_t2_r, sc01_05_t2_r, sc01_06_t2_r)), na.rm=T),
concept_par2_t2 = rowMeans(across(c(sc01_02_t2_r, sc01_09_t2_r, sc01_10_t2_r, sc01_07_t2_r)), na.rm=T),
concept_par3_t2 = rowMeans(across(c(sc01_03_t2_r, sc01_08_t2_r, sc01_12_t2_r, sc01_11_t2)), na.rm=T))
# replace NaN with regular NA
df_sbsa3_wide_wb <- df_sbsa3_wide_wb %>%
mutate_all(~ifelse(is.nan(.), NA, .))# all time points
df_sbsa3_wide_pers <- df_sbsa3 %>%
arrange(pid, time) %>%
select(pid, group, time, starts_with(c("bf05", "bf06"))) %>%
mutate(valid=1) %>%
pivot_wider(names_from = time,
names_sep = "_t",
values_from = c(valid, starts_with(c("bf05", "bf06"))))
# colnames(df_sbsa3_wide_pers)Check CFA item loadings at T1
cfa_extra_curr <- '
# Define the latent factors
extra_curr1 =~ NA*bf05_01_t1 + lambda1*bf05_01_t1 + lambda2*bf05_06_t1 + lambda3*bf05_11_t1 + lambda4*bf05_16_t1 + lambda5*bf05_21_t1 + lambda6*bf05_26_t1 + lambda7*bf05_31_t1 + lambda8*bf05_36_t1 + lambda9*bf05_41_t1 + lambda10*bf05_46_t1 + lambda11*bf05_51_t1 + lambda12*bf05_56_t1
# Intercepts
bf05_01_t1 ~ i1*1
bf05_06_t1 ~ 1
bf05_11_t1 ~ 1
bf05_16_t1 ~ 1
bf05_21_t1 ~ 1
bf05_26_t1 ~ 1
bf05_31_t1 ~ 1
bf05_36_t1 ~ 1
bf05_41_t1 ~ 1
bf05_46_t1 ~ 1
bf05_51_t1 ~ 1
bf05_56_t1 ~ 1
# Unique Variances
bf05_01_t1 ~~ bf05_01_t1
bf05_06_t1 ~~ bf05_06_t1
bf05_11_t1 ~~ bf05_11_t1
bf05_16_t1 ~~ bf05_16_t1
bf05_21_t1 ~~ bf05_21_t1
bf05_26_t1 ~~ bf05_26_t1
bf05_31_t1 ~~ bf05_31_t1
bf05_36_t1 ~~ bf05_36_t1
bf05_41_t1 ~~ bf05_41_t1
bf05_46_t1 ~~ bf05_46_t1
bf05_51_t1 ~~ bf05_51_t1
bf05_56_t1 ~~ bf05_56_t1
# Latent Variable Means
extra_curr1 ~ 0*1
# Latent Variable Variances and Covariance
extra_curr1 ~~ 1*extra_curr1
'
fit_cfa_extra_curr <- cfa(cfa_extra_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_cfa_extra_curr, fit.measures = TRUE)lavaan 0.6.15 ended normally after 15 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 529 530
Number of missing patterns 1
Model Test User Model:
Test statistic 586.241
Degrees of freedom 54
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 2319.986
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.764
Tucker-Lewis Index (TLI) 0.711
Robust Comparative Fit Index (CFI) 0.764
Robust Tucker-Lewis Index (TLI) 0.711
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -9321.205
Loglikelihood unrestricted model (H1) -9028.085
Akaike (AIC) 18714.411
Bayesian (BIC) 18868.166
Sample-size adjusted Bayesian (SABIC) 18753.892
Root Mean Square Error of Approximation:
RMSEA 0.136
90 Percent confidence interval - lower 0.127
90 Percent confidence interval - upper 0.147
P-value H_0: RMSEA <= 0.050 0.000
P-value H_0: RMSEA >= 0.080 1.000
Robust RMSEA 0.136
90 Percent confidence interval - lower 0.127
90 Percent confidence interval - upper 0.147
P-value H_0: Robust RMSEA <= 0.050 0.000
P-value H_0: Robust RMSEA >= 0.080 1.000
Standardized Root Mean Square Residual:
SRMR 0.078
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
extra_curr1 =~
b05_01_ (lmb1) 1.037 0.051 20.495 0.000
b05_06_ (lmb2) 0.561 0.052 10.811 0.000
b05_11_ (lmb3) -0.485 0.051 -9.583 0.000
b05_16_ (lmb4) -0.790 0.051 -15.586 0.000
b05_21_ (lmb5) 0.736 0.055 13.446 0.000
b05_26_ (lmb6) -0.611 0.057 -10.709 0.000
b05_31_ (lmb7) -0.769 0.047 -16.183 0.000
b05_36_ (lmb8) -0.535 0.048 -11.050 0.000
b05_41_ (lmb9) 0.716 0.047 15.145 0.000
b05_46_ (lm10) 0.893 0.055 16.239 0.000
b05_51_ (lm11) -0.606 0.052 -11.600 0.000
b05_56_ (lm12) 0.709 0.045 15.859 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf05_01_1 (i1) 3.017 0.057 52.480 0.000
.bf05_06_1 3.223 0.052 62.522 0.000
.bf05_11_1 2.595 0.050 52.215 0.000
.bf05_16_1 3.798 0.053 71.597 0.000
.bf05_21_1 2.885 0.056 51.442 0.000
.bf05_26_1 2.909 0.056 51.598 0.000
.bf05_31_1 3.947 0.050 78.443 0.000
.bf05_36_1 2.966 0.048 61.476 0.000
.bf05_41_1 3.002 0.049 60.942 0.000
.bf05_46_1 3.115 0.058 53.301 0.000
.bf05_51_1 3.059 0.052 58.645 0.000
.bf05_56_1 3.323 0.047 70.392 0.000
extr_crr1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf05_01_t1 0.672 0.054 12.502 0.000
.bf05_06_t1 1.091 0.070 15.561 0.000
.bf05_11_t1 1.072 0.068 15.733 0.000
.bf05_16_t1 0.865 0.061 14.247 0.000
.bf05_21_t1 1.121 0.074 15.049 0.000
.bf05_26_t1 1.309 0.084 15.523 0.000
.bf05_31_t1 0.748 0.053 14.180 0.000
.bf05_36_t1 0.945 0.061 15.528 0.000
.bf05_41_t1 0.772 0.053 14.467 0.000
.bf05_46_t1 1.009 0.071 14.261 0.000
.bf05_51_t1 1.071 0.070 15.353 0.000
.bf05_56_t1 0.676 0.047 14.326 0.000
extra_curr1 1.000 tidy(fit_cfa_extra_curr) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 extra_curr1 =~ bf05_01_t1 =~ lambda1 1.04 0.0506 0.785
2 extra_curr1 =~ bf05_46_t1 =~ lambda10 0.893 0.0550 0.664
3 extra_curr1 =~ bf05_16_t1 =~ lambda4 -0.790 0.0507 -0.647
4 extra_curr1 =~ bf05_31_t1 =~ lambda7 -0.769 0.0475 -0.664
5 extra_curr1 =~ bf05_21_t1 =~ lambda5 0.736 0.0548 0.571
6 extra_curr1 =~ bf05_41_t1 =~ lambda9 0.716 0.0472 0.632
7 extra_curr1 =~ bf05_56_t1 =~ lambda12 0.709 0.0447 0.653
8 extra_curr1 =~ bf05_26_t1 =~ lambda6 -0.611 0.0570 -0.471
9 extra_curr1 =~ bf05_51_t1 =~ lambda11 -0.606 0.0523 -0.506
10 extra_curr1 =~ bf05_06_t1 =~ lambda2 0.561 0.0519 0.473
11 extra_curr1 =~ bf05_36_t1 =~ lambda8 -0.535 0.0484 -0.482
12 extra_curr1 =~ bf05_11_t1 =~ lambda3 -0.485 0.0506 -0.424Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf05_31_t1_r = bf05_31_t1,
bf05_16_t1_r = bf05_16_t1,
bf05_26_t1_r = bf05_26_t1,
bf05_51_t1_r = bf05_51_t1,
bf05_36_t1_r = bf05_36_t1,
bf05_11_t1_r = bf05_11_t1,
bf05_31_t2_r = bf05_31_t2,
bf05_16_t2_r = bf05_16_t2,
bf05_26_t2_r = bf05_26_t2,
bf05_51_t2_r = bf05_51_t2,
bf05_36_t2_r = bf05_36_t2,
bf05_11_t2_r = bf05_11_t2) %>%
mutate(across(c(bf05_31_t1_r, bf05_16_t1_r, bf05_26_t1_r, bf05_51_t1_r, bf05_36_t1_r, bf05_11_t1_r,
bf05_31_t2_r, bf05_16_t2_r, bf05_26_t2_r, bf05_51_t2_r, bf05_36_t2_r, bf05_11_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(extra_curr_par1_t1 = rowMeans(across(c(bf05_01_t1, bf05_41_t1, bf05_51_t1_r, bf05_11_t1_r)), na.rm=T),
extra_curr_par2_t1 = rowMeans(across(c(bf05_46_t1, bf05_21_t1, bf05_26_t1_r, bf05_36_t1_r)), na.rm=T),
extra_curr_par3_t1 = rowMeans(across(c(bf05_16_t1_r, bf05_31_t1_r, bf05_56_t1, bf05_06_t1)), na.rm=T),
extra_curr_par1_t2 = rowMeans(across(c(bf05_01_t2, bf05_41_t2, bf05_51_t2_r, bf05_11_t2_r)), na.rm=T),
extra_curr_par2_t2 = rowMeans(across(c(bf05_46_t2, bf05_21_t2, bf05_26_t2_r, bf05_36_t2_r)), na.rm=T),
extra_curr_par3_t2 = rowMeans(across(c(bf05_16_t2_r, bf05_31_t2_r, bf05_56_t2, bf05_06_t2)), na.rm=T))Check CFA item loadings at T1
cfa_extra_ideal <- '
# Define the latent factors
extra_ideal1 =~ NA*bf06_01_t1 + lambda1*bf06_01_t1 + lambda2*bf06_06_t1 + lambda3*bf06_11_t1 + lambda4*bf06_16_t1 + lambda5*bf06_21_t1 + lambda6*bf06_26_t1 + lambda7*bf06_31_t1 + lambda8*bf06_36_t1 + lambda9*bf06_41_t1 + lambda10*bf06_46_t1 + lambda11*bf06_51_t1 + lambda12*bf06_56_t1
# Intercepts
bf06_01_t1 ~ i1*1
bf06_06_t1 ~ 1
bf06_11_t1 ~ 1
bf06_16_t1 ~ 1
bf06_21_t1 ~ 1
bf06_26_t1 ~ 1
bf06_31_t1 ~ 1
bf06_36_t1 ~ 1
bf06_41_t1 ~ 1
bf06_46_t1 ~ 1
bf06_51_t1 ~ 1
bf06_56_t1 ~ 1
# Unique Variances
bf06_01_t1 ~~ bf06_01_t1
bf06_06_t1 ~~ bf06_06_t1
bf06_11_t1 ~~ bf06_11_t1
bf06_16_t1 ~~ bf06_16_t1
bf06_21_t1 ~~ bf06_21_t1
bf06_26_t1 ~~ bf06_26_t1
bf06_31_t1 ~~ bf06_31_t1
bf06_36_t1 ~~ bf06_36_t1
bf06_41_t1 ~~ bf06_41_t1
bf06_46_t1 ~~ bf06_46_t1
bf06_51_t1 ~~ bf06_51_t1
bf06_56_t1 ~~ bf06_56_t1
# Latent Variable Means
extra_ideal1 ~ 0*1
# Latent Variable Variances and Covariance
extra_ideal1 ~~ 1*extra_ideal1
'
fit_cfa_extra_ideal <- cfa(cfa_extra_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_cfa_extra_ideal, fit.measures = TRUE)lavaan 0.6.15 ended normally after 16 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 174 175
Number of missing patterns 1
Model Test User Model:
Test statistic 108.405
Degrees of freedom 54
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 365.082
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.818
Tucker-Lewis Index (TLI) 0.778
Robust Comparative Fit Index (CFI) 0.818
Robust Tucker-Lewis Index (TLI) 0.778
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -2843.995
Loglikelihood unrestricted model (H1) -2789.793
Akaike (AIC) 5759.991
Bayesian (BIC) 5873.717
Sample-size adjusted Bayesian (SABIC) 5759.718
Root Mean Square Error of Approximation:
RMSEA 0.076
90 Percent confidence interval - lower 0.055
90 Percent confidence interval - upper 0.097
P-value H_0: RMSEA <= 0.050 0.022
P-value H_0: RMSEA >= 0.080 0.396
Robust RMSEA 0.076
90 Percent confidence interval - lower 0.055
90 Percent confidence interval - upper 0.097
P-value H_0: Robust RMSEA <= 0.050 0.022
P-value H_0: Robust RMSEA >= 0.080 0.396
Standardized Root Mean Square Residual:
SRMR 0.063
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
extra_ideal1 =~
b06_01_ (lmb1) 0.549 0.070 7.808 0.000
b06_06_ (lmb2) 0.318 0.075 4.225 0.000
b06_11_ (lmb3) -0.261 0.109 -2.392 0.017
b06_16_ (lmb4) -0.630 0.098 -6.443 0.000
b06_21_ (lmb5) 0.512 0.086 5.935 0.000
b06_26_ (lmb6) -0.397 0.071 -5.598 0.000
b06_31_ (lmb7) -0.717 0.086 -8.315 0.000
b06_36_ (lmb8) -0.478 0.095 -5.025 0.000
b06_41_ (lmb9) 0.429 0.066 6.458 0.000
b06_46_ (lm10) 0.375 0.081 4.612 0.000
b06_51_ (lm11) -0.440 0.090 -4.883 0.000
b06_56_ (lm12) 0.300 0.070 4.292 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf06_01_1 (i1) 4.092 0.068 60.062 0.000
.bf06_06_1 4.161 0.068 60.865 0.000
.bf06_11_1 2.109 0.096 21.975 0.000
.bf06_16_1 2.943 0.092 31.911 0.000
.bf06_21_1 3.960 0.079 49.891 0.000
.bf06_26_1 1.632 0.066 24.865 0.000
.bf06_31_1 2.287 0.084 27.230 0.000
.bf06_36_1 2.190 0.087 25.082 0.000
.bf06_41_1 4.402 0.062 70.564 0.000
.bf06_46_1 3.730 0.074 50.326 0.000
.bf06_51_1 2.356 0.082 28.572 0.000
.bf06_56_1 4.218 0.063 66.535 0.000
extra_dl1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf06_01_t1 0.506 0.066 7.676 0.000
.bf06_06_t1 0.712 0.080 8.936 0.000
.bf06_11_t1 1.535 0.167 9.193 0.000
.bf06_16_t1 1.082 0.130 8.310 0.000
.bf06_21_t1 0.834 0.100 8.341 0.000
.bf06_26_t1 0.592 0.069 8.579 0.000
.bf06_31_t1 0.713 0.098 7.248 0.000
.bf06_36_t1 1.098 0.125 8.749 0.000
.bf06_41_t1 0.493 0.060 8.262 0.000
.bf06_46_t1 0.815 0.092 8.842 0.000
.bf06_51_t1 0.990 0.113 8.778 0.000
.bf06_56_t1 0.609 0.068 8.916 0.000
extra_ideal1 1.000 tidy(fit_cfa_extra_ideal) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 extra_ideal1 =~ bf06_31_t1 =~ lambda7 -0.717 0.0863 -0.647
2 extra_ideal1 =~ bf06_16_t1 =~ lambda4 -0.630 0.0978 -0.518
3 extra_ideal1 =~ bf06_01_t1 =~ lambda1 0.549 0.0703 0.611
4 extra_ideal1 =~ bf06_21_t1 =~ lambda5 0.512 0.0863 0.489
5 extra_ideal1 =~ bf06_36_t1 =~ lambda8 -0.478 0.0951 -0.415
6 extra_ideal1 =~ bf06_51_t1 =~ lambda11 -0.440 0.0901 -0.404
7 extra_ideal1 =~ bf06_41_t1 =~ lambda9 0.429 0.0664 0.521
8 extra_ideal1 =~ bf06_26_t1 =~ lambda6 -0.397 0.0709 -0.458
9 extra_ideal1 =~ bf06_46_t1 =~ lambda10 0.375 0.0814 0.384
10 extra_ideal1 =~ bf06_06_t1 =~ lambda2 0.318 0.0753 0.353
11 extra_ideal1 =~ bf06_56_t1 =~ lambda12 0.300 0.0699 0.359
12 extra_ideal1 =~ bf06_11_t1 =~ lambda3 -0.261 0.109 -0.206Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf06_31_t1_r = bf06_31_t1,
bf06_16_t1_r = bf06_16_t1,
bf06_26_t1_r = bf06_26_t1,
bf06_51_t1_r = bf06_51_t1,
bf06_36_t1_r = bf06_36_t1,
bf06_11_t1_r = bf06_11_t1,
bf06_31_t2_r = bf06_31_t2,
bf06_16_t2_r = bf06_16_t2,
bf06_26_t2_r = bf06_26_t2,
bf06_51_t2_r = bf06_51_t2,
bf06_36_t2_r = bf06_36_t2,
bf06_11_t2_r = bf06_11_t2) %>%
mutate(across(c(bf06_31_t1_r, bf06_16_t1_r, bf06_26_t1_r, bf06_51_t1_r, bf06_36_t1_r, bf06_11_t1_r,
bf06_31_t2_r, bf06_16_t2_r, bf06_26_t2_r, bf06_51_t2_r, bf06_36_t2_r, bf06_11_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(extra_ideal_par1_t1 = rowMeans(across(c(bf06_31_t1_r, bf06_51_t1_r, bf06_46_t1, bf06_11_t1_r)), na.rm=T),
extra_ideal_par2_t1 = rowMeans(across(c(bf06_16_t1_r, bf06_36_t1_r, bf06_26_t1_r, bf06_56_t1)), na.rm=T),
extra_ideal_par3_t1 = rowMeans(across(c(bf06_01_t1, bf06_21_t1, bf06_41_t1, bf06_06_t1)), na.rm=T),
extra_ideal_par1_t2 = rowMeans(across(c(bf06_31_t2_r, bf06_51_t2_r, bf06_46_t2, bf06_11_t2_r)), na.rm=T),
extra_ideal_par2_t2 = rowMeans(across(c(bf06_16_t2_r, bf06_36_t2_r, bf06_26_t2_r, bf06_56_t2)), na.rm=T),
extra_ideal_par3_t2 = rowMeans(across(c(bf06_01_t2, bf06_21_t2, bf06_41_t2, bf06_06_t2)), na.rm=T))Check CFA item loadings at T1
cfa_agree_curr <- '
# Define the latent factors
agree_curr1 =~ NA*bf05_02_t1 + lambda1*bf05_02_t1 + lambda2*bf05_07_t1 + lambda3*bf05_12_t1 + lambda4*bf05_17_t1 + lambda5*bf05_22_t1 + lambda6*bf05_27_t1 + lambda7*bf05_32_t1 + lambda8*bf05_37_t1 + lambda9*bf05_42_t1 + lambda10*bf05_47_t1 + lambda11*bf05_52_t1 + lambda12*bf05_57_t1
# Intercepts
bf05_02_t1 ~ i1*1
bf05_07_t1 ~ 1
bf05_12_t1 ~ 1
bf05_17_t1 ~ 1
bf05_22_t1 ~ 1
bf05_27_t1 ~ 1
bf05_32_t1 ~ 1
bf05_37_t1 ~ 1
bf05_42_t1 ~ 1
bf05_47_t1 ~ 1
bf05_52_t1 ~ 1
bf05_57_t1 ~ 1
# Unique Variances
bf05_02_t1 ~~ bf05_02_t1
bf05_07_t1 ~~ bf05_07_t1
bf05_12_t1 ~~ bf05_12_t1
bf05_17_t1 ~~ bf05_17_t1
bf05_22_t1 ~~ bf05_22_t1
bf05_27_t1 ~~ bf05_27_t1
bf05_32_t1 ~~ bf05_32_t1
bf05_37_t1 ~~ bf05_37_t1
bf05_42_t1 ~~ bf05_42_t1
bf05_47_t1 ~~ bf05_47_t1
bf05_52_t1 ~~ bf05_52_t1
bf05_57_t1 ~~ bf05_57_t1
# Latent Variable Means
agree_curr1 ~ 0*1
# Latent Variable Variances and Covariance
agree_curr1 ~~ 1*agree_curr1
'
fit_cfa_agree_curr <- cfa(cfa_agree_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_cfa_agree_curr, fit.measures = TRUE)lavaan 0.6.15 ended normally after 20 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 529 530
Number of missing patterns 1
Model Test User Model:
Test statistic 308.188
Degrees of freedom 54
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 1434.558
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.814
Tucker-Lewis Index (TLI) 0.773
Robust Comparative Fit Index (CFI) 0.814
Robust Tucker-Lewis Index (TLI) 0.773
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -8687.839
Loglikelihood unrestricted model (H1) -8533.745
Akaike (AIC) 17447.677
Bayesian (BIC) 17601.433
Sample-size adjusted Bayesian (SABIC) 17487.159
Root Mean Square Error of Approximation:
RMSEA 0.094
90 Percent confidence interval - lower 0.084
90 Percent confidence interval - upper 0.105
P-value H_0: RMSEA <= 0.050 0.000
P-value H_0: RMSEA >= 0.080 0.990
Robust RMSEA 0.094
90 Percent confidence interval - lower 0.084
90 Percent confidence interval - upper 0.105
P-value H_0: Robust RMSEA <= 0.050 0.000
P-value H_0: Robust RMSEA >= 0.080 0.990
Standardized Root Mean Square Residual:
SRMR 0.061
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
agree_curr1 =~
b05_02_ (lmb1) 0.489 0.036 13.767 0.000
b05_07_ (lmb2) 0.404 0.029 14.046 0.000
b05_12_ (lmb3) -0.626 0.050 -12.413 0.000
b05_17_ (lmb4) -0.242 0.070 -3.472 0.001
b05_22_ (lmb5) -0.545 0.053 -10.203 0.000
b05_27_ (lmb6) 0.615 0.051 11.961 0.000
b05_32_ (lmb7) 0.504 0.041 12.371 0.000
b05_37_ (lmb8) -0.674 0.054 -12.558 0.000
b05_42_ (lmb9) -0.368 0.054 -6.781 0.000
b05_47_ (lm10) -0.714 0.053 -13.414 0.000
b05_52_ (lm11) 0.420 0.032 13.050 0.000
b05_57_ (lm12) 0.611 0.051 12.068 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf05_02_1 (i1) 4.321 0.036 121.608 0.000
.bf05_07_1 4.505 0.029 156.371 0.000
.bf05_12_1 2.802 0.050 56.576 0.000
.bf05_17_1 2.699 0.064 42.458 0.000
.bf05_22_1 2.227 0.051 43.517 0.000
.bf05_27_1 3.637 0.050 72.600 0.000
.bf05_32_1 4.021 0.040 100.182 0.000
.bf05_37_1 2.527 0.053 47.877 0.000
.bf05_42_1 3.376 0.050 67.691 0.000
.bf05_47_1 2.457 0.053 46.331 0.000
.bf05_52_1 4.282 0.032 135.111 0.000
.bf05_57_1 3.302 0.049 67.014 0.000
agre_crr1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf05_02_t1 0.429 0.030 14.208 0.000
.bf05_07_t1 0.276 0.020 14.020 0.000
.bf05_12_t1 0.906 0.062 14.651 0.000
.bf05_17_t1 2.080 0.129 16.165 0.000
.bf05_22_t1 1.088 0.071 15.248 0.000
.bf05_27_t1 0.950 0.064 14.751 0.000
.bf05_32_t1 0.598 0.041 14.738 0.000
.bf05_37_t1 1.020 0.070 14.593 0.000
.bf05_42_t1 1.181 0.075 15.802 0.000
.bf05_47_t1 0.979 0.068 14.363 0.000
.bf05_52_t1 0.355 0.025 14.309 0.000
.bf05_57_t1 0.911 0.062 14.646 0.000
agree_curr1 1.000 tidy(fit_cfa_agree_curr) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 agree_curr1 =~ bf05_47_t1 =~ lambda10 -0.714 0.0532 -0.585
2 agree_curr1 =~ bf05_37_t1 =~ lambda8 -0.674 0.0537 -0.555
3 agree_curr1 =~ bf05_12_t1 =~ lambda3 -0.626 0.0504 -0.549
4 agree_curr1 =~ bf05_27_t1 =~ lambda6 0.615 0.0514 0.533
5 agree_curr1 =~ bf05_57_t1 =~ lambda12 0.611 0.0506 0.539
6 agree_curr1 =~ bf05_22_t1 =~ lambda5 -0.545 0.0534 -0.463
7 agree_curr1 =~ bf05_32_t1 =~ lambda7 0.504 0.0407 0.546
8 agree_curr1 =~ bf05_02_t1 =~ lambda1 0.489 0.0355 0.598
9 agree_curr1 =~ bf05_52_t1 =~ lambda11 0.420 0.0322 0.576
10 agree_curr1 =~ bf05_07_t1 =~ lambda2 0.404 0.0288 0.610
11 agree_curr1 =~ bf05_42_t1 =~ lambda9 -0.368 0.0542 -0.320
12 agree_curr1 =~ bf05_17_t1 =~ lambda4 -0.242 0.0696 -0.165Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf05_47_t1_r = bf05_47_t1,
bf05_37_t1_r = bf05_37_t1,
bf05_22_t1_r = bf05_22_t1,
bf05_12_t1_r = bf05_12_t1,
bf05_42_t1_r = bf05_42_t1,
bf05_17_t1_r = bf05_17_t1,
bf05_47_t2_r = bf05_47_t2,
bf05_37_t2_r = bf05_37_t2,
bf05_22_t2_r = bf05_22_t2,
bf05_12_t2_r = bf05_12_t2,
bf05_42_t2_r = bf05_42_t2,
bf05_17_t2_r = bf05_17_t2) %>%
mutate(across(c(bf05_47_t1_r, bf05_37_t1_r, bf05_22_t1_r, bf05_12_t1_r, bf05_42_t1_r, bf05_17_t1_r,
bf05_47_t2_r, bf05_37_t2_r, bf05_22_t2_r, bf05_12_t2_r, bf05_42_t2_r, bf05_17_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(agree_curr_par1_t1 = rowMeans(across(c(bf05_47_t1_r, bf05_22_t1_r, bf05_52_t1, bf05_17_t1_r)), na.rm=T),
agree_curr_par2_t1 = rowMeans(across(c(bf05_37_t1_r, bf05_57_t1, bf05_02_t1, bf05_42_t1_r)), na.rm=T),
agree_curr_par3_t1 = rowMeans(across(c(bf05_12_t1_r, bf05_27_t1, bf05_32_t1, bf05_07_t1)), na.rm=T),
agree_curr_par1_t2 = rowMeans(across(c(bf05_47_t2_r, bf05_22_t2_r, bf05_52_t2, bf05_17_t2_r)), na.rm=T),
agree_curr_par2_t2 = rowMeans(across(c(bf05_37_t2_r, bf05_57_t2, bf05_02_t2, bf05_42_t2_r)), na.rm=T),
agree_curr_par3_t2 = rowMeans(across(c(bf05_12_t2_r, bf05_27_t2, bf05_32_t2, bf05_07_t2)), na.rm=T))Check CFA item loadings at T1
cfa_agree_ideal <- '
# Define the latent factors
agree_ideal1 =~ NA*bf06_02_t1 + lambda1*bf06_02_t1 + lambda2*bf06_07_t1 + lambda3*bf06_12_t1 + lambda4*bf06_17_t1 + lambda5*bf06_22_t1 + lambda6*bf06_27_t1 + lambda7*bf06_32_t1 + lambda8*bf06_37_t1 + lambda9*bf06_42_t1 + lambda10*bf06_47_t1 + lambda11*bf06_52_t1 + lambda12*bf06_57_t1
# Intercepts
bf06_02_t1 ~ i1*1
bf06_07_t1 ~ 1
bf06_12_t1 ~ 1
bf06_17_t1 ~ 1
bf06_22_t1 ~ 1
bf06_27_t1 ~ 1
bf06_32_t1 ~ 1
bf06_37_t1 ~ 1
bf06_42_t1 ~ 1
bf06_47_t1 ~ 1
bf06_52_t1 ~ 1
bf06_57_t1 ~ 1
# Unique Variances
bf06_02_t1 ~~ bf06_02_t1
bf06_07_t1 ~~ bf06_07_t1
bf06_12_t1 ~~ bf06_12_t1
bf06_17_t1 ~~ bf06_17_t1
bf06_22_t1 ~~ bf06_22_t1
bf06_27_t1 ~~ bf06_27_t1
bf06_32_t1 ~~ bf06_32_t1
bf06_37_t1 ~~ bf06_37_t1
bf06_42_t1 ~~ bf06_42_t1
bf06_47_t1 ~~ bf06_47_t1
bf06_52_t1 ~~ bf06_52_t1
bf06_57_t1 ~~ bf06_57_t1
# Latent Variable Means
agree_ideal1 ~ 0*1
# Latent Variable Variances and Covariance
agree_ideal1 ~~ 1*agree_ideal1
'
fit_cfa_agree_ideal <- cfa(cfa_agree_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_cfa_agree_ideal, fit.measures = TRUE)lavaan 0.6.15 ended normally after 23 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 174 175
Number of missing patterns 1
Model Test User Model:
Test statistic 81.870
Degrees of freedom 54
P-value (Chi-square) 0.009
Model Test Baseline Model:
Test statistic 369.885
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.908
Tucker-Lewis Index (TLI) 0.888
Robust Comparative Fit Index (CFI) 0.908
Robust Tucker-Lewis Index (TLI) 0.888
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -2727.336
Loglikelihood unrestricted model (H1) -2686.401
Akaike (AIC) 5526.672
Bayesian (BIC) 5640.398
Sample-size adjusted Bayesian (SABIC) 5526.400
Root Mean Square Error of Approximation:
RMSEA 0.054
90 Percent confidence interval - lower 0.028
90 Percent confidence interval - upper 0.077
P-value H_0: RMSEA <= 0.050 0.359
P-value H_0: RMSEA >= 0.080 0.032
Robust RMSEA 0.054
90 Percent confidence interval - lower 0.028
90 Percent confidence interval - upper 0.077
P-value H_0: Robust RMSEA <= 0.050 0.359
P-value H_0: Robust RMSEA >= 0.080 0.032
Standardized Root Mean Square Residual:
SRMR 0.054
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
agree_ideal1 =~
b06_02_ (lmb1) 0.474 0.070 6.823 0.000
b06_07_ (lmb2) 0.281 0.044 6.360 0.000
b06_12_ (lmb3) -0.536 0.082 -6.496 0.000
b06_17_ (lmb4) -0.129 0.135 -0.951 0.341
b06_22_ (lmb5) -0.450 0.080 -5.626 0.000
b06_27_ (lmb6) 0.620 0.085 7.340 0.000
b06_32_ (lmb7) 0.366 0.079 4.602 0.000
b06_37_ (lmb8) -0.504 0.077 -6.580 0.000
b06_42_ (lmb9) -0.417 0.094 -4.445 0.000
b06_47_ (lm10) -0.510 0.083 -6.158 0.000
b06_52_ (lm11) 0.335 0.054 6.239 0.000
b06_57_ (lm12) 0.574 0.076 7.587 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf06_02_1 (i1) 4.299 0.067 64.507 0.000
.bf06_07_1 4.724 0.042 113.168 0.000
.bf06_12_1 1.983 0.079 25.232 0.000
.bf06_17_1 2.678 0.120 22.318 0.000
.bf06_22_1 1.707 0.075 22.775 0.000
.bf06_27_1 3.862 0.082 47.201 0.000
.bf06_32_1 4.270 0.073 58.321 0.000
.bf06_37_1 1.718 0.073 23.550 0.000
.bf06_42_1 2.690 0.085 31.464 0.000
.bf06_47_1 1.810 0.078 23.175 0.000
.bf06_52_1 4.557 0.050 90.415 0.000
.bf06_57_1 3.730 0.074 50.631 0.000
agree_dl1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf06_02_t1 0.548 0.066 8.271 0.000
.bf06_07_t1 0.224 0.027 8.380 0.000
.bf06_12_t1 0.787 0.094 8.395 0.000
.bf06_17_t1 2.489 0.267 9.311 0.000
.bf06_22_t1 0.774 0.090 8.652 0.000
.bf06_27_t1 0.780 0.097 8.050 0.000
.bf06_32_t1 0.799 0.090 8.904 0.000
.bf06_37_t1 0.673 0.081 8.350 0.000
.bf06_42_t1 1.098 0.123 8.896 0.000
.bf06_47_t1 0.802 0.095 8.473 0.000
.bf06_52_t1 0.330 0.039 8.375 0.000
.bf06_57_t1 0.615 0.077 7.944 0.000
agree_ideal1 1.000 tidy(fit_cfa_agree_ideal) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 agree_ideal1 =~ bf06_27_t1 =~ lambda6 0.620 0.0845 0.575
2 agree_ideal1 =~ bf06_57_t1 =~ lambda12 0.574 0.0757 0.591
3 agree_ideal1 =~ bf06_12_t1 =~ lambda3 -0.536 0.0825 -0.517
4 agree_ideal1 =~ bf06_47_t1 =~ lambda10 -0.510 0.0828 -0.495
5 agree_ideal1 =~ bf06_37_t1 =~ lambda8 -0.504 0.0766 -0.523
6 agree_ideal1 =~ bf06_02_t1 =~ lambda1 0.474 0.0695 0.540
7 agree_ideal1 =~ bf06_22_t1 =~ lambda5 -0.450 0.0801 -0.456
8 agree_ideal1 =~ bf06_42_t1 =~ lambda9 -0.417 0.0937 -0.370
9 agree_ideal1 =~ bf06_32_t1 =~ lambda7 0.366 0.0795 0.379
10 agree_ideal1 =~ bf06_52_t1 =~ lambda11 0.335 0.0537 0.504
11 agree_ideal1 =~ bf06_07_t1 =~ lambda2 0.281 0.0442 0.510
12 agree_ideal1 =~ bf06_17_t1 =~ lambda4 -0.129 0.135 -0.0812Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf06_47_t1_r = bf06_47_t1,
bf06_37_t1_r = bf06_37_t1,
bf06_22_t1_r = bf06_22_t1,
bf06_12_t1_r = bf06_12_t1,
bf06_42_t1_r = bf06_42_t1,
bf06_17_t1_r = bf06_17_t1,
bf06_47_t2_r = bf06_47_t2,
bf06_37_t2_r = bf06_37_t2,
bf06_22_t2_r = bf06_22_t2,
bf06_12_t2_r = bf06_12_t2,
bf06_42_t2_r = bf06_42_t2,
bf06_17_t2_r = bf06_17_t2) %>%
mutate(across(c(bf06_47_t1_r, bf06_37_t1_r, bf06_22_t1_r, bf06_12_t1_r, bf06_42_t1_r, bf06_17_t1_r,
bf06_47_t2_r, bf06_37_t2_r, bf06_22_t2_r, bf06_12_t2_r, bf06_42_t2_r, bf06_17_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(agree_ideal_par1_t1 = rowMeans(across(c(bf06_27_t1, bf06_02_t1, bf06_32_t1, bf06_17_t1_r)), na.rm=T),
agree_ideal_par2_t1 = rowMeans(across(c(bf06_57_t1, bf06_37_t1_r, bf06_42_t1_r, bf06_07_t1)), na.rm=T),
agree_ideal_par3_t1 = rowMeans(across(c(bf06_12_t1_r, bf06_47_t1_r, bf06_22_t1_r, bf06_52_t1)), na.rm=T),
agree_ideal_par1_t2 = rowMeans(across(c(bf06_27_t2, bf06_02_t2, bf06_32_t2, bf06_17_t2_r)), na.rm=T),
agree_ideal_par2_t2 = rowMeans(across(c(bf06_57_t2, bf06_37_t2_r, bf06_42_t2_r, bf06_07_t2)), na.rm=T),
agree_ideal_par3_t2 = rowMeans(across(c(bf06_12_t2_r, bf06_47_t2_r, bf06_22_t2_r, bf06_52_t2)), na.rm=T))Check CFA item loadings at T1
cfa_consc_curr <- '
# Define the latent factors
consc_curr1 =~ NA*bf05_03_t1 + lambda1*bf05_03_t1 + lambda2*bf05_08_t1 + lambda3*bf05_13_t1 + lambda4*bf05_18_t1 + lambda5*bf05_23_t1 + lambda6*bf05_28_t1 + lambda7*bf05_33_t1 + lambda8*bf05_38_t1 + lambda9*bf05_43_t1 + lambda10*bf05_48_t1 + lambda11*bf05_53_t1 + lambda12*bf05_58_t1
# Intercepts
bf05_03_t1 ~ i1*1
bf05_08_t1 ~ 1
bf05_13_t1 ~ 1
bf05_18_t1 ~ 1
bf05_23_t1 ~ 1
bf05_28_t1 ~ 1
bf05_33_t1 ~ 1
bf05_38_t1 ~ 1
bf05_43_t1 ~ 1
bf05_48_t1 ~ 1
bf05_53_t1 ~ 1
bf05_58_t1 ~ 1
# Unique Variances
bf05_03_t1 ~~ bf05_03_t1
bf05_08_t1 ~~ bf05_08_t1
bf05_13_t1 ~~ bf05_13_t1
bf05_18_t1 ~~ bf05_18_t1
bf05_23_t1 ~~ bf05_23_t1
bf05_28_t1 ~~ bf05_28_t1
bf05_33_t1 ~~ bf05_33_t1
bf05_38_t1 ~~ bf05_38_t1
bf05_43_t1 ~~ bf05_43_t1
bf05_48_t1 ~~ bf05_48_t1
bf05_53_t1 ~~ bf05_53_t1
bf05_58_t1 ~~ bf05_58_t1
# Latent Variable Means
consc_curr1 ~ 0*1
# Latent Variable Variances and Covariance
consc_curr1 ~~ 1*consc_curr1
'
fit_cfa_consc_curr <- cfa(cfa_consc_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_cfa_consc_curr, fit.measures = TRUE)lavaan 0.6.15 ended normally after 16 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 529 530
Number of missing patterns 1
Model Test User Model:
Test statistic 473.373
Degrees of freedom 54
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 2427.188
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.822
Tucker-Lewis Index (TLI) 0.783
Robust Comparative Fit Index (CFI) 0.822
Robust Tucker-Lewis Index (TLI) 0.783
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -8913.589
Loglikelihood unrestricted model (H1) -8676.903
Akaike (AIC) 17899.178
Bayesian (BIC) 18052.934
Sample-size adjusted Bayesian (SABIC) 17938.660
Root Mean Square Error of Approximation:
RMSEA 0.121
90 Percent confidence interval - lower 0.111
90 Percent confidence interval - upper 0.131
P-value H_0: RMSEA <= 0.050 0.000
P-value H_0: RMSEA >= 0.080 1.000
Robust RMSEA 0.121
90 Percent confidence interval - lower 0.111
90 Percent confidence interval - upper 0.131
P-value H_0: Robust RMSEA <= 0.050 0.000
P-value H_0: Robust RMSEA >= 0.080 1.000
Standardized Root Mean Square Residual:
SRMR 0.067
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
consc_curr1 =~
b05_03_ (lmb1) 1.035 0.052 19.907 0.000
b05_08_ (lmb2) 0.730 0.051 14.218 0.000
b05_13_ (lmb3) -0.367 0.046 -7.969 0.000
b05_18_ (lmb4) -0.814 0.048 -17.100 0.000
b05_23_ (lmb5) 0.779 0.054 14.445 0.000
b05_28_ (lmb6) 0.670 0.050 13.471 0.000
b05_33_ (lmb7) -0.923 0.047 -19.565 0.000
b05_38_ (lmb8) -0.631 0.041 -15.256 0.000
b05_43_ (lmb9) -0.421 0.039 -10.775 0.000
b05_48_ (lm10) 0.781 0.048 16.299 0.000
b05_53_ (lm11) -0.661 0.045 -14.551 0.000
b05_58_ (lm12) 0.641 0.052 12.300 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf05_03_1 (i1) 2.896 0.059 49.361 0.000
.bf05_08_1 3.465 0.054 64.742 0.000
.bf05_13_1 3.728 0.045 83.125 0.000
.bf05_18_1 3.614 0.052 70.129 0.000
.bf05_23_1 3.331 0.056 59.159 0.000
.bf05_28_1 2.907 0.051 56.674 0.000
.bf05_33_1 3.416 0.053 64.433 0.000
.bf05_38_1 3.722 0.044 85.516 0.000
.bf05_43_1 4.070 0.039 104.755 0.000
.bf05_48_1 2.087 0.051 40.584 0.000
.bf05_53_1 3.784 0.047 80.181 0.000
.bf05_58_1 2.699 0.053 50.831 0.000
cnsc_crr1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf05_03_t1 0.750 0.057 13.059 0.000
.bf05_08_t1 0.982 0.065 15.089 0.000
.bf05_13_t1 0.929 0.058 15.950 0.000
.bf05_18_t1 0.743 0.052 14.274 0.000
.bf05_23_t1 1.070 0.071 15.003 0.000
.bf05_28_t1 0.944 0.062 15.225 0.000
.bf05_33_t1 0.634 0.048 13.277 0.000
.bf05_38_t1 0.604 0.041 14.697 0.000
.bf05_43_t1 0.622 0.040 15.596 0.000
.bf05_48_t1 0.789 0.054 14.615 0.000
.bf05_53_t1 0.742 0.050 14.806 0.000
.bf05_58_t1 1.081 0.070 15.457 0.000
consc_curr1 1.000 tidy(fit_cfa_consc_curr) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 consc_curr1 =~ bf05_03_t1 =~ lambda1 1.03 0.0520 0.767
2 consc_curr1 =~ bf05_33_t1 =~ lambda7 -0.923 0.0472 -0.757
3 consc_curr1 =~ bf05_18_t1 =~ lambda4 -0.814 0.0476 -0.687
4 consc_curr1 =~ bf05_48_t1 =~ lambda10 0.781 0.0479 0.660
5 consc_curr1 =~ bf05_23_t1 =~ lambda5 0.779 0.0539 0.602
6 consc_curr1 =~ bf05_08_t1 =~ lambda2 0.730 0.0514 0.593
7 consc_curr1 =~ bf05_28_t1 =~ lambda6 0.670 0.0497 0.568
8 consc_curr1 =~ bf05_53_t1 =~ lambda11 -0.661 0.0454 -0.609
9 consc_curr1 =~ bf05_58_t1 =~ lambda12 0.641 0.0521 0.525
10 consc_curr1 =~ bf05_38_t1 =~ lambda8 -0.631 0.0414 -0.631
11 consc_curr1 =~ bf05_43_t1 =~ lambda9 -0.421 0.0390 -0.471
12 consc_curr1 =~ bf05_13_t1 =~ lambda3 -0.367 0.0461 -0.356Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf05_03_t1_r = bf05_03_t1,
bf05_23_t1_r = bf05_23_t1,
bf05_48_t1_r = bf05_48_t1,
bf05_08_t1_r = bf05_08_t1,
bf05_28_t1_r = bf05_28_t1,
bf05_58_t1_r = bf05_58_t1,
bf05_03_t2_r = bf05_03_t2,
bf05_23_t2_r = bf05_23_t2,
bf05_48_t2_r = bf05_48_t2,
bf05_08_t2_r = bf05_08_t2,
bf05_28_t2_r = bf05_28_t2,
bf05_58_t2_r = bf05_58_t2) %>%
mutate(across(c(bf05_03_t1_r, bf05_23_t1_r, bf05_48_t1_r, bf05_08_t1_r, bf05_28_t1_r, bf05_58_t1_r,
bf05_03_t2_r, bf05_23_t2_r, bf05_48_t2_r, bf05_08_t2_r, bf05_28_t2_r, bf05_58_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(consc_curr_par1_t1 = rowMeans(across(c(bf05_03_t1_r, bf05_08_t1_r, bf05_58_t1_r, bf05_13_t1)), na.rm=T),
consc_curr_par2_t1 = rowMeans(across(c(bf05_33_t1, bf05_23_t1_r, bf05_53_t1, bf05_43_t1)), na.rm=T),
consc_curr_par3_t1 = rowMeans(across(c(bf05_18_t1, bf05_48_t1_r, bf05_28_t1_r, bf05_38_t1)), na.rm=T),
consc_curr_par1_t2 = rowMeans(across(c(bf05_03_t2_r, bf05_08_t2_r, bf05_58_t2_r, bf05_13_t2)), na.rm=T),
consc_curr_par2_t2 = rowMeans(across(c(bf05_33_t2, bf05_23_t2_r, bf05_53_t2, bf05_43_t2)), na.rm=T),
consc_curr_par3_t2 = rowMeans(across(c(bf05_18_t2, bf05_48_t2_r, bf05_28_t2_r, bf05_38_t2)), na.rm=T))Check CFA item loadings at T1
cfa_consc_ideal <- '
# Define the latent factors
consc_ideal1 =~ NA*bf06_03_t1 + lambda1*bf06_03_t1 + lambda2*bf06_08_t1 + lambda3*bf06_13_t1 + lambda4*bf06_18_t1 + lambda5*bf06_23_t1 + lambda6*bf06_28_t1 + lambda7*bf06_33_t1 + lambda8*bf06_38_t1 + lambda9*bf06_43_t1 + lambda10*bf06_48_t1 + lambda11*bf06_53_t1 + lambda12*bf06_58_t1
# Intercepts
bf06_03_t1 ~ i1*1
bf06_08_t1 ~ 1
bf06_13_t1 ~ 1
bf06_18_t1 ~ 1
bf06_23_t1 ~ 1
bf06_28_t1 ~ 1
bf06_33_t1 ~ 1
bf06_38_t1 ~ 1
bf06_43_t1 ~ 1
bf06_48_t1 ~ 1
bf06_53_t1 ~ 1
bf06_58_t1 ~ 1
# Unique Variances
bf06_03_t1 ~~ bf06_03_t1
bf06_08_t1 ~~ bf06_08_t1
bf06_13_t1 ~~ bf06_13_t1
bf06_18_t1 ~~ bf06_18_t1
bf06_23_t1 ~~ bf06_23_t1
bf06_28_t1 ~~ bf06_28_t1
bf06_33_t1 ~~ bf06_33_t1
bf06_38_t1 ~~ bf06_38_t1
bf06_43_t1 ~~ bf06_43_t1
bf06_48_t1 ~~ bf06_48_t1
bf06_53_t1 ~~ bf06_53_t1
bf06_58_t1 ~~ bf06_58_t1
# Latent Variable Means
consc_ideal1 ~ 0*1
# Latent Variable Variances and Covariance
consc_ideal1 ~~ 1*consc_ideal1
'
fit_cfa_consc_ideal <- cfa(cfa_consc_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_cfa_consc_ideal, fit.measures = TRUE)lavaan 0.6.15 ended normally after 38 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 174 175
Number of missing patterns 2
Model Test User Model:
Test statistic 85.313
Degrees of freedom 54
P-value (Chi-square) 0.004
Model Test Baseline Model:
Test statistic 403.053
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.907
Tucker-Lewis Index (TLI) 0.886
Robust Comparative Fit Index (CFI) 0.907
Robust Tucker-Lewis Index (TLI) 0.887
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -2492.025
Loglikelihood unrestricted model (H1) -2449.369
Akaike (AIC) 5056.051
Bayesian (BIC) 5169.777
Sample-size adjusted Bayesian (SABIC) 5055.778
Root Mean Square Error of Approximation:
RMSEA 0.058
90 Percent confidence interval - lower 0.033
90 Percent confidence interval - upper 0.080
P-value H_0: RMSEA <= 0.050 0.277
P-value H_0: RMSEA >= 0.080 0.052
Robust RMSEA 0.058
90 Percent confidence interval - lower 0.033
90 Percent confidence interval - upper 0.080
P-value H_0: Robust RMSEA <= 0.050 0.278
P-value H_0: Robust RMSEA >= 0.080 0.052
Standardized Root Mean Square Residual:
SRMR 0.059
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
consc_ideal1 =~
b06_03_ (lmb1) 0.422 0.094 4.477 0.000
b06_08_ (lmb2) 0.415 0.071 5.875 0.000
b06_13_ (lmb3) -0.188 0.092 -2.049 0.040
b06_18_ (lmb4) -0.337 0.066 -5.103 0.000
b06_23_ (lmb5) 0.521 0.071 7.346 0.000
b06_28_ (lmb6) 0.423 0.086 4.944 0.000
b06_33_ (lmb7) -0.358 0.061 -5.859 0.000
b06_38_ (lmb8) -0.378 0.046 -8.194 0.000
b06_43_ (lmb9) -0.240 0.064 -3.735 0.000
b06_48_ (lm10) 0.525 0.056 9.382 0.000
b06_53_ (lm11) -0.508 0.054 -9.483 0.000
b06_58_ (lm12) 0.409 0.074 5.561 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf06_03_1 (i1) 1.759 0.088 20.010 0.000
.bf06_08_1 1.638 0.068 24.244 0.000
.bf06_13_1 4.259 0.083 51.458 0.000
.bf06_18_1 4.379 0.062 70.457 0.000
.bf06_23_1 1.540 0.070 22.088 0.000
.bf06_28_1 1.977 0.080 24.704 0.000
.bf06_33_1 4.425 0.058 76.044 0.000
.bf06_38_1 4.667 0.046 100.975 0.000
.bf06_43_1 4.500 0.059 76.232 0.000
.bf06_48_1 1.377 0.058 23.799 0.000
.bf06_53_1 4.546 0.055 81.965 0.000
.bf06_58_1 1.747 0.070 25.084 0.000
consc_dl1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf06_03_t1 1.166 0.130 8.995 0.000
.bf06_08_t1 0.622 0.071 8.727 0.000
.bf06_13_t1 1.157 0.125 9.259 0.000
.bf06_18_t1 0.558 0.063 8.868 0.000
.bf06_23_t1 0.575 0.069 8.284 0.000
.bf06_28_t1 0.936 0.105 8.887 0.000
.bf06_33_t1 0.461 0.053 8.708 0.000
.bf06_38_t1 0.229 0.029 7.947 0.000
.bf06_43_t1 0.549 0.060 9.091 0.000
.bf06_48_t1 0.305 0.042 7.334 0.000
.bf06_53_t1 0.277 0.038 7.300 0.000
.bf06_58_t1 0.676 0.077 8.759 0.000
consc_ideal1 1.000 tidy(fit_cfa_consc_ideal) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 consc_ideal1 =~ bf06_48_t1 =~ lambda10 0.525 0.0560 0.689
2 consc_ideal1 =~ bf06_23_t1 =~ lambda5 0.521 0.0709 0.566
3 consc_ideal1 =~ bf06_53_t1 =~ lambda11 -0.508 0.0536 -0.695
4 consc_ideal1 =~ bf06_28_t1 =~ lambda6 0.423 0.0856 0.401
5 consc_ideal1 =~ bf06_03_t1 =~ lambda1 0.422 0.0942 0.364
6 consc_ideal1 =~ bf06_08_t1 =~ lambda2 0.415 0.0706 0.466
7 consc_ideal1 =~ bf06_58_t1 =~ lambda12 0.409 0.0736 0.446
8 consc_ideal1 =~ bf06_38_t1 =~ lambda8 -0.378 0.0461 -0.620
9 consc_ideal1 =~ bf06_33_t1 =~ lambda7 -0.358 0.0610 -0.466
10 consc_ideal1 =~ bf06_18_t1 =~ lambda4 -0.337 0.0661 -0.412
11 consc_ideal1 =~ bf06_43_t1 =~ lambda9 -0.240 0.0641 -0.308
12 consc_ideal1 =~ bf06_13_t1 =~ lambda3 -0.188 0.0915 -0.172Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf06_03_t1_r = bf06_03_t1,
bf06_23_t1_r = bf06_23_t1,
bf06_48_t1_r = bf06_48_t1,
bf06_08_t1_r = bf06_08_t1,
bf06_28_t1_r = bf06_28_t1,
bf06_58_t1_r = bf06_58_t1,
bf06_03_t2_r = bf06_03_t2,
bf06_23_t2_r = bf06_23_t2,
bf06_48_t2_r = bf06_48_t2,
bf06_08_t2_r = bf06_08_t2,
bf06_28_t2_r = bf06_28_t2,
bf06_58_t2_r = bf06_58_t2) %>%
mutate(across(c(bf06_03_t1_r, bf06_23_t1_r, bf06_48_t1_r, bf06_08_t1_r, bf06_28_t1_r, bf06_58_t1_r,
bf06_03_t2_r, bf06_23_t2_r, bf06_48_t2_r, bf06_08_t2_r, bf06_28_t2_r, bf06_58_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(consc_ideal_par1_t1 = rowMeans(across(c(bf06_48_t1_r, bf06_08_t1_r, bf06_33_t1, bf06_13_t1)), na.rm=T),
consc_ideal_par2_t1 = rowMeans(across(c(bf06_23_t1_r, bf06_03_t1_r, bf06_38_t1, bf06_43_t1)), na.rm=T),
consc_ideal_par3_t1 = rowMeans(across(c(bf06_53_t1, bf06_28_t1_r, bf06_58_t1_r, bf06_18_t1)), na.rm=T),
consc_ideal_par1_t2 = rowMeans(across(c(bf06_48_t2_r, bf06_08_t2_r, bf06_33_t2, bf06_13_t2)), na.rm=T),
consc_ideal_par2_t2 = rowMeans(across(c(bf06_23_t2_r, bf06_03_t2_r, bf06_38_t2, bf06_43_t2)), na.rm=T),
consc_ideal_par3_t2 = rowMeans(across(c(bf06_53_t2, bf06_28_t2_r, bf06_58_t2_r, bf06_18_t2)), na.rm=T))Check CFA item loadings at T1
cfa_neuro_curr <- '
#| warning: false
# Define the latent factors
neuro_curr1 =~ NA*bf05_04_t1 + lambda1*bf05_04_t1 + lambda2*bf05_09_t1 + lambda3*bf05_14_t1 + lambda4*bf05_19_t1 + lambda5*bf05_24_t1 + lambda6*bf05_29_t1 + lambda7*bf05_34_t1 + lambda8*bf05_39_t1 + lambda9*bf05_44_t1 + lambda10*bf05_49_t1 + lambda11*bf05_54_t1 + lambda12*bf05_59_t1
# Intercepts
bf05_04_t1 ~ i1*1
bf05_09_t1 ~ 1
bf05_14_t1 ~ 1
bf05_19_t1 ~ 1
bf05_24_t1 ~ 1
bf05_29_t1 ~ 1
bf05_34_t1 ~ 1
bf05_39_t1 ~ 1
bf05_44_t1 ~ 1
bf05_49_t1 ~ 1
bf05_54_t1 ~ 1
bf05_59_t1 ~ 1
# Unique Variances
bf05_04_t1 ~~ bf05_04_t1
bf05_09_t1 ~~ bf05_09_t1
bf05_14_t1 ~~ bf05_14_t1
bf05_19_t1 ~~ bf05_19_t1
bf05_24_t1 ~~ bf05_24_t1
bf05_29_t1 ~~ bf05_29_t1
bf05_34_t1 ~~ bf05_34_t1
bf05_39_t1 ~~ bf05_39_t1
bf05_44_t1 ~~ bf05_44_t1
bf05_49_t1 ~~ bf05_49_t1
bf05_54_t1 ~~ bf05_54_t1
bf05_59_t1 ~~ bf05_59_t1
# Latent Variable Means
neuro_curr1 ~ 0*1
# Latent Variable Variances and Covariance
neuro_curr1 ~~ 1*neuro_curr1
'
fit_cfa_neuro_curr <- cfa(cfa_neuro_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")Warning in lav_data_full(data = data, group = group, cluster = cluster, : lavaan WARNING: some cases are empty and will be ignored:
191summary(fit_cfa_neuro_curr, fit.measures = TRUE)lavaan 0.6.15 ended normally after 14 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 529 530
Number of missing patterns 1
Model Test User Model:
Test statistic 468.967
Degrees of freedom 54
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 3276.139
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.871
Tucker-Lewis Index (TLI) 0.842
Robust Comparative Fit Index (CFI) 0.871
Robust Tucker-Lewis Index (TLI) 0.842
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -8968.375
Loglikelihood unrestricted model (H1) -8733.891
Akaike (AIC) 18008.749
Bayesian (BIC) 18162.505
Sample-size adjusted Bayesian (SABIC) 18048.231
Root Mean Square Error of Approximation:
RMSEA 0.121
90 Percent confidence interval - lower 0.111
90 Percent confidence interval - upper 0.131
P-value H_0: RMSEA <= 0.050 0.000
P-value H_0: RMSEA >= 0.080 1.000
Robust RMSEA 0.121
90 Percent confidence interval - lower 0.111
90 Percent confidence interval - upper 0.131
P-value H_0: Robust RMSEA <= 0.050 0.000
P-value H_0: Robust RMSEA >= 0.080 1.000
Standardized Root Mean Square Residual:
SRMR 0.052
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
neuro_curr1 =~
b05_04_ (lmb1) 0.924 0.049 18.742 0.000
b05_09_ (lmb2) 0.781 0.050 15.715 0.000
b05_14_ (lmb3) -0.892 0.051 -17.320 0.000
b05_19_ (lmb4) -0.619 0.045 -13.822 0.000
b05_24_ (lmb5) 0.737 0.051 14.464 0.000
b05_29_ (lmb6) 1.006 0.047 21.320 0.000
b05_34_ (lmb7) -0.895 0.047 -18.926 0.000
b05_39_ (lmb8) -0.950 0.048 -19.949 0.000
b05_44_ (lmb9) 0.784 0.045 17.367 0.000
b05_49_ (lm10) 0.726 0.052 13.877 0.000
b05_54_ (lm11) -0.992 0.050 -19.931 0.000
b05_59_ (lm12) -0.874 0.051 -17.150 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf05_04_1 (i1) 2.820 0.055 50.879 0.000
.bf05_09_1 2.994 0.054 55.829 0.000
.bf05_14_1 3.121 0.057 54.964 0.000
.bf05_19_1 3.550 0.047 75.245 0.000
.bf05_24_1 3.257 0.054 60.205 0.000
.bf05_29_1 3.057 0.055 55.388 0.000
.bf05_34_1 3.732 0.053 69.884 0.000
.bf05_39_1 3.234 0.054 59.500 0.000
.bf05_44_1 3.355 0.050 67.515 0.000
.bf05_49_1 2.397 0.055 43.476 0.000
.bf05_54_1 3.144 0.057 55.339 0.000
.bf05_59_1 3.040 0.056 54.262 0.000
neur_crr1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf05_04_t1 0.771 0.053 14.533 0.000
.bf05_09_t1 0.911 0.060 15.269 0.000
.bf05_14_t1 0.911 0.061 14.973 0.000
.bf05_19_t1 0.794 0.051 15.559 0.000
.bf05_24_t1 1.005 0.065 15.463 0.000
.bf05_29_t1 0.599 0.044 13.666 0.000
.bf05_34_t1 0.707 0.048 14.569 0.000
.bf05_39_t1 0.661 0.047 13.953 0.000
.bf05_44_t1 0.692 0.047 14.854 0.000
.bf05_49_t1 1.080 0.070 15.544 0.000
.bf05_54_t1 0.724 0.052 13.981 0.000
.bf05_59_t1 0.896 0.060 14.968 0.000
neuro_curr1 1.000 tidy(fit_cfa_neuro_curr) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 neuro_curr1 =~ bf05_29_t1 =~ lambda6 1.01 0.0472 0.793
2 neuro_curr1 =~ bf05_54_t1 =~ lambda11 -0.992 0.0498 -0.759
3 neuro_curr1 =~ bf05_39_t1 =~ lambda8 -0.950 0.0476 -0.760
4 neuro_curr1 =~ bf05_04_t1 =~ lambda1 0.924 0.0493 0.725
5 neuro_curr1 =~ bf05_34_t1 =~ lambda7 -0.895 0.0473 -0.729
6 neuro_curr1 =~ bf05_14_t1 =~ lambda3 -0.892 0.0515 -0.683
7 neuro_curr1 =~ bf05_59_t1 =~ lambda12 -0.874 0.0510 -0.678
8 neuro_curr1 =~ bf05_44_t1 =~ lambda9 0.784 0.0451 0.686
9 neuro_curr1 =~ bf05_09_t1 =~ lambda2 0.781 0.0497 0.633
10 neuro_curr1 =~ bf05_24_t1 =~ lambda5 0.737 0.0510 0.593
11 neuro_curr1 =~ bf05_49_t1 =~ lambda10 0.726 0.0523 0.573
12 neuro_curr1 =~ bf05_19_t1 =~ lambda4 -0.619 0.0448 -0.571Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf05_29_t1_r = bf05_29_t1,
bf05_04_t1_r = bf05_04_t1,
bf05_24_t1_r = bf05_24_t1,
bf05_44_t1_r = bf05_44_t1,
bf05_09_t1_r = bf05_09_t1,
bf05_49_t1_r = bf05_49_t1,
bf05_29_t2_r = bf05_29_t2,
bf05_04_t2_r = bf05_04_t2,
bf05_24_t2_r = bf05_24_t2,
bf05_44_t2_r = bf05_44_t2,
bf05_09_t2_r = bf05_09_t2,
bf05_49_t2_r = bf05_49_t2) %>%
mutate(across(c(bf05_29_t1_r, bf05_04_t1_r, bf05_24_t1_r, bf05_44_t1_r, bf05_09_t1_r, bf05_49_t1_r,
bf05_29_t2_r, bf05_04_t2_r, bf05_24_t2_r, bf05_44_t2_r, bf05_09_t2_r, bf05_49_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(neuro_curr_par1_t1 = rowMeans(across(c(bf05_29_t1_r, bf05_14_t1, bf05_09_t1_r, bf05_19_t1)), na.rm=T),
neuro_curr_par2_t1 = rowMeans(across(c(bf05_54_t1, bf05_34_t1, bf05_44_t1_r, bf05_49_t1_r)), na.rm=T),
neuro_curr_par3_t1 = rowMeans(across(c(bf05_39_t1, bf05_04_t1_r, bf05_59_t1, bf05_24_t1_r)), na.rm=T),
neuro_curr_par1_t2 = rowMeans(across(c(bf05_29_t2_r, bf05_14_t2, bf05_09_t2_r, bf05_19_t2)), na.rm=T),
neuro_curr_par2_t2 = rowMeans(across(c(bf05_54_t2, bf05_34_t2, bf05_44_t2_r, bf05_49_t2_r)), na.rm=T),
neuro_curr_par3_t2 = rowMeans(across(c(bf05_39_t2, bf05_04_t2_r, bf05_59_t2, bf05_24_t2_r)), na.rm=T))Check CFA item loadings at T1
cfa_neuro_ideal <- '
# Define the latent factors
neuro_ideal1 =~ NA*bf06_04_t1 + lambda1*bf06_04_t1 + lambda2*bf06_09_t1 + lambda3*bf06_14_t1 + lambda4*bf06_19_t1 + lambda5*bf06_24_t1 + lambda6*bf06_29_t1 + lambda7*bf06_34_t1 + lambda8*bf06_39_t1 + lambda9*bf06_44_t1 + lambda10*bf06_49_t1 + lambda11*bf06_54_t1 + lambda12*bf06_59_t1
# Intercepts
bf06_04_t1 ~ i1*1
bf06_09_t1 ~ 1
bf06_14_t1 ~ 1
bf06_19_t1 ~ 1
bf06_24_t1 ~ 1
bf06_29_t1 ~ 1
bf06_34_t1 ~ 1
bf06_39_t1 ~ 1
bf06_44_t1 ~ 1
bf06_49_t1 ~ 1
bf06_54_t1 ~ 1
bf06_59_t1 ~ 1
# Unique Variances
bf06_04_t1 ~~ bf06_04_t1
bf06_09_t1 ~~ bf06_09_t1
bf06_14_t1 ~~ bf06_14_t1
bf06_19_t1 ~~ bf06_19_t1
bf06_24_t1 ~~ bf06_24_t1
bf06_29_t1 ~~ bf06_29_t1
bf06_34_t1 ~~ bf06_34_t1
bf06_39_t1 ~~ bf06_39_t1
bf06_44_t1 ~~ bf06_44_t1
bf06_49_t1 ~~ bf06_49_t1
bf06_54_t1 ~~ bf06_54_t1
bf06_59_t1 ~~ bf06_59_t1
# Latent Variable Means
neuro_ideal1 ~ 0*1
# Latent Variable Variances and Covariance
neuro_ideal1 ~~ 1*neuro_ideal1
'
fit_cfa_neuro_ideal <- cfa(cfa_neuro_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_cfa_neuro_ideal, fit.measures = TRUE)lavaan 0.6.15 ended normally after 20 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 174 175
Number of missing patterns 1
Model Test User Model:
Test statistic 103.156
Degrees of freedom 54
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 389.315
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.848
Tucker-Lewis Index (TLI) 0.814
Robust Comparative Fit Index (CFI) 0.848
Robust Tucker-Lewis Index (TLI) 0.814
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -2571.036
Loglikelihood unrestricted model (H1) -2519.458
Akaike (AIC) 5214.072
Bayesian (BIC) 5327.798
Sample-size adjusted Bayesian (SABIC) 5213.800
Root Mean Square Error of Approximation:
RMSEA 0.072
90 Percent confidence interval - lower 0.051
90 Percent confidence interval - upper 0.093
P-value H_0: RMSEA <= 0.050 0.044
P-value H_0: RMSEA >= 0.080 0.289
Robust RMSEA 0.072
90 Percent confidence interval - lower 0.051
90 Percent confidence interval - upper 0.093
P-value H_0: Robust RMSEA <= 0.050 0.044
P-value H_0: Robust RMSEA >= 0.080 0.289
Standardized Root Mean Square Residual:
SRMR 0.062
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
neuro_ideal1 =~
b06_04_ (lmb1) 0.216 0.060 3.608 0.000
b06_09_ (lmb2) 0.298 0.059 5.090 0.000
b06_14_ (lmb3) -0.416 0.075 -5.548 0.000
b06_19_ (lmb4) -0.335 0.067 -5.009 0.000
b06_24_ (lmb5) 0.232 0.065 3.600 0.000
b06_29_ (lmb6) 0.329 0.088 3.733 0.000
b06_34_ (lmb7) -0.601 0.074 -8.141 0.000
b06_39_ (lmb8) -0.638 0.067 -9.460 0.000
b06_44_ (lmb9) 0.311 0.065 4.780 0.000
b06_49_ (lm10) 0.480 0.116 4.130 0.000
b06_54_ (lm11) -0.525 0.060 -8.790 0.000
b06_59_ (lm12) -0.507 0.077 -6.549 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf06_04_1 (i1) 4.563 0.055 83.323 0.000
.bf06_09_1 4.425 0.055 80.928 0.000
.bf06_14_1 1.661 0.070 23.676 0.000
.bf06_19_1 1.954 0.062 31.618 0.000
.bf06_24_1 4.638 0.059 79.061 0.000
.bf06_29_1 4.299 0.080 53.640 0.000
.bf06_34_1 1.787 0.072 24.654 0.000
.bf06_39_1 1.632 0.068 23.963 0.000
.bf06_44_1 4.351 0.060 72.304 0.000
.bf06_49_1 3.580 0.107 33.572 0.000
.bf06_54_1 1.460 0.060 24.304 0.000
.bf06_59_1 1.931 0.074 25.987 0.000
neuro_dl1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf06_04_t1 0.475 0.052 9.089 0.000
.bf06_09_t1 0.431 0.049 8.834 0.000
.bf06_14_t1 0.683 0.079 8.681 0.000
.bf06_19_t1 0.552 0.063 8.788 0.000
.bf06_24_t1 0.545 0.060 9.078 0.000
.bf06_29_t1 1.009 0.111 9.051 0.000
.bf06_34_t1 0.553 0.073 7.595 0.000
.bf06_39_t1 0.401 0.060 6.690 0.000
.bf06_44_t1 0.533 0.060 8.867 0.000
.bf06_49_t1 1.749 0.194 9.006 0.000
.bf06_54_t1 0.352 0.048 7.402 0.000
.bf06_59_t1 0.703 0.083 8.441 0.000
neuro_ideal1 1.000 tidy(fit_cfa_neuro_ideal) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 neuro_ideal1 =~ bf06_39_t1 =~ lambda8 -0.638 0.0674 -0.710
2 neuro_ideal1 =~ bf06_34_t1 =~ lambda7 -0.601 0.0738 -0.629
3 neuro_ideal1 =~ bf06_54_t1 =~ lambda11 -0.525 0.0598 -0.663
4 neuro_ideal1 =~ bf06_59_t1 =~ lambda12 -0.507 0.0775 -0.518
5 neuro_ideal1 =~ bf06_49_t1 =~ lambda10 0.480 0.116 0.341
6 neuro_ideal1 =~ bf06_14_t1 =~ lambda3 -0.416 0.0750 -0.450
7 neuro_ideal1 =~ bf06_19_t1 =~ lambda4 -0.335 0.0670 -0.411
8 neuro_ideal1 =~ bf06_29_t1 =~ lambda6 0.329 0.0882 0.311
9 neuro_ideal1 =~ bf06_44_t1 =~ lambda9 0.311 0.0651 0.392
10 neuro_ideal1 =~ bf06_09_t1 =~ lambda2 0.298 0.0585 0.413
11 neuro_ideal1 =~ bf06_24_t1 =~ lambda5 0.232 0.0645 0.300
12 neuro_ideal1 =~ bf06_04_t1 =~ lambda1 0.216 0.0600 0.300Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf06_29_t1_r = bf06_29_t1,
bf06_04_t1_r = bf06_04_t1,
bf06_24_t1_r = bf06_24_t1,
bf06_44_t1_r = bf06_44_t1,
bf06_09_t1_r = bf06_09_t1,
bf06_49_t1_r = bf06_49_t1,
bf06_29_t2_r = bf06_29_t2,
bf06_04_t2_r = bf06_04_t2,
bf06_24_t2_r = bf06_24_t2,
bf06_44_t2_r = bf06_44_t2,
bf06_09_t2_r = bf06_09_t2,
bf06_49_t2_r = bf06_49_t2) %>%
mutate(across(c(bf06_29_t1_r, bf06_04_t1_r, bf06_24_t1_r, bf06_44_t1_r, bf06_09_t1_r, bf06_49_t1_r,
bf06_29_t2_r, bf06_04_t2_r, bf06_24_t2_r, bf06_44_t2_r, bf06_09_t2_r, bf06_49_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(neuro_ideal_par1_t1 = rowMeans(across(c(bf06_39_t1, bf06_14_t1, bf06_44_t1_r, bf06_04_t1_r)), na.rm=T),
neuro_ideal_par2_t1 = rowMeans(across(c(bf06_34_t1, bf06_49_t1, bf06_29_t1_r, bf06_24_t1_r)), na.rm=T),
neuro_ideal_par3_t1 = rowMeans(across(c(bf06_54_t1, bf06_59_t1, bf06_19_t1, bf06_09_t1_r)), na.rm=T),
neuro_ideal_par1_t2 = rowMeans(across(c(bf06_39_t2, bf06_14_t2, bf06_44_t2_r, bf06_04_t2_r)), na.rm=T),
neuro_ideal_par2_t2 = rowMeans(across(c(bf06_34_t2, bf06_49_t2, bf06_29_t2_r, bf06_24_t2_r)), na.rm=T),
neuro_ideal_par3_t2 = rowMeans(across(c(bf06_54_t2, bf06_59_t2, bf06_19_t2, bf06_09_t2_r)), na.rm=T))Check CFA item loadings at T1
cfa_openn_curr <- '
# Define the latent factors
openn_curr1 =~ NA*bf05_05_t1 + lambda1*bf05_05_t1 + lambda2*bf05_10_t1 + lambda3*bf05_15_t1 + lambda4*bf05_20_t1 + lambda5*bf05_25_t1 + lambda6*bf05_30_t1 + lambda7*bf05_35_t1 + lambda8*bf05_40_t1 + lambda9*bf05_45_t1 + lambda10*bf05_50_t1 + lambda11*bf05_55_t1 + lambda12*bf05_60_t1
# Intercepts
bf05_05_t1 ~ i1*1
bf05_10_t1 ~ 1
bf05_15_t1 ~ 1
bf05_20_t1 ~ 1
bf05_25_t1 ~ 1
bf05_30_t1 ~ 1
bf05_35_t1 ~ 1
bf05_40_t1 ~ 1
bf05_45_t1 ~ 1
bf05_50_t1 ~ 1
bf05_55_t1 ~ 1
bf05_60_t1 ~ 1
# Unique Variances
bf05_05_t1 ~~ bf05_05_t1
bf05_10_t1 ~~ bf05_10_t1
bf05_15_t1 ~~ bf05_15_t1
bf05_20_t1 ~~ bf05_20_t1
bf05_25_t1 ~~ bf05_25_t1
bf05_30_t1 ~~ bf05_30_t1
bf05_35_t1 ~~ bf05_35_t1
bf05_40_t1 ~~ bf05_40_t1
bf05_45_t1 ~~ bf05_45_t1
bf05_50_t1 ~~ bf05_50_t1
bf05_55_t1 ~~ bf05_55_t1
bf05_60_t1 ~~ bf05_60_t1
# Latent Variable Means
openn_curr1 ~ 0*1
# Latent Variable Variances and Covariance
openn_curr1 ~~ 1*openn_curr1
'
fit_cfa_openn_curr <- cfa(cfa_openn_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_cfa_openn_curr, fit.measures = TRUE)lavaan 0.6.15 ended normally after 38 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 529 530
Number of missing patterns 3
Model Test User Model:
Test statistic 485.860
Degrees of freedom 54
P-value (Chi-square) 0.000
Model Test Baseline Model:
Test statistic 1752.038
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.744
Tucker-Lewis Index (TLI) 0.687
Robust Comparative Fit Index (CFI) 0.744
Robust Tucker-Lewis Index (TLI) 0.687
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -9023.572
Loglikelihood unrestricted model (H1) -8780.642
Akaike (AIC) 18119.144
Bayesian (BIC) 18272.900
Sample-size adjusted Bayesian (SABIC) 18158.626
Root Mean Square Error of Approximation:
RMSEA 0.123
90 Percent confidence interval - lower 0.113
90 Percent confidence interval - upper 0.133
P-value H_0: RMSEA <= 0.050 0.000
P-value H_0: RMSEA >= 0.080 1.000
Robust RMSEA 0.123
90 Percent confidence interval - lower 0.113
90 Percent confidence interval - upper 0.133
P-value H_0: Robust RMSEA <= 0.050 0.000
P-value H_0: Robust RMSEA >= 0.080 1.000
Standardized Root Mean Square Residual:
SRMR 0.071
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
openn_curr1 =~
b05_05_ (lmb1) 0.264 0.062 4.250 0.000
b05_10_ (lmb2) -0.472 0.039 -12.041 0.000
b05_15_ (lmb3) -0.496 0.045 -10.959 0.000
b05_20_ (lmb4) -0.718 0.052 -13.924 0.000
b05_25_ (lmb5) 0.508 0.052 9.740 0.000
b05_30_ (lmb6) 0.763 0.054 14.191 0.000
b05_35_ (lmb7) -0.737 0.044 -16.590 0.000
b05_40_ (lmb8) -0.396 0.044 -9.033 0.000
b05_45_ (lmb9) 0.600 0.049 12.235 0.000
b05_50_ (lm10) 0.733 0.057 12.840 0.000
b05_55_ (lm11) 0.656 0.051 12.799 0.000
b05_60_ (lm12) -0.612 0.045 -13.592 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf05_05_1 (i1) 2.947 0.057 51.273 0.000
.bf05_10_1 4.257 0.039 110.013 0.000
.bf05_15_1 3.756 0.043 86.840 0.000
.bf05_20_1 3.917 0.050 77.742 0.000
.bf05_25_1 2.146 0.050 42.723 0.000
.bf05_30_1 2.514 0.054 46.215 0.000
.bf05_35_1 3.964 0.045 87.351 0.000
.bf05_40_1 4.030 0.042 96.362 0.000
.bf05_45_1 1.991 0.048 41.146 0.000
.bf05_50_1 2.378 0.057 41.878 0.000
.bf05_55_1 2.461 0.051 48.058 0.000
.bf05_60_1 3.603 0.044 81.225 0.000
opnn_crr1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf05_05_t1 1.678 0.104 16.129 0.000
.bf05_10_t1 0.570 0.038 14.974 0.000
.bf05_15_t1 0.744 0.050 14.897 0.000
.bf05_20_t1 0.828 0.062 13.363 0.000
.bf05_25_t1 1.076 0.070 15.472 0.000
.bf05_30_t1 0.982 0.069 14.206 0.000
.bf05_35_t1 0.547 0.045 12.077 0.000
.bf05_40_t1 0.769 0.049 15.585 0.000
.bf05_45_t1 0.878 0.059 14.851 0.000
.bf05_50_t1 1.168 0.080 14.680 0.000
.bf05_55_t1 0.955 0.064 14.810 0.000
.bf05_60_t1 0.667 0.048 13.910 0.000
openn_curr1 1.000 tidy(fit_cfa_openn_curr) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 openn_curr1 =~ bf05_30_t1 =~ lambda6 0.763 0.0538 0.610
2 openn_curr1 =~ bf05_35_t1 =~ lambda7 -0.737 0.0444 -0.706
3 openn_curr1 =~ bf05_50_t1 =~ lambda10 0.733 0.0571 0.561
4 openn_curr1 =~ bf05_20_t1 =~ lambda4 -0.718 0.0515 -0.619
5 openn_curr1 =~ bf05_55_t1 =~ lambda11 0.656 0.0512 0.557
6 openn_curr1 =~ bf05_60_t1 =~ lambda12 -0.612 0.0450 -0.600
7 openn_curr1 =~ bf05_45_t1 =~ lambda9 0.600 0.0490 0.539
8 openn_curr1 =~ bf05_25_t1 =~ lambda5 0.508 0.0522 0.440
9 openn_curr1 =~ bf05_15_t1 =~ lambda3 -0.496 0.0452 -0.498
10 openn_curr1 =~ bf05_10_t1 =~ lambda2 -0.472 0.0392 -0.530
11 openn_curr1 =~ bf05_40_t1 =~ lambda8 -0.396 0.0438 -0.411
12 openn_curr1 =~ bf05_05_t1 =~ lambda1 0.264 0.0622 0.200Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf05_50_t1_r = bf05_50_t1,
bf05_55_t1_r = bf05_55_t1,
bf05_30_t1_r = bf05_30_t1,
bf05_25_t1_r = bf05_25_t1,
bf05_45_t1_r = bf05_45_t1,
bf05_05_t1_r = bf05_05_t1,
bf05_50_t2_r = bf05_50_t2,
bf05_55_t2_r = bf05_55_t2,
bf05_30_t2_r = bf05_30_t2,
bf05_25_t2_r = bf05_25_t2,
bf05_45_t2_r = bf05_45_t2,
bf05_05_t2_r = bf05_05_t2) %>%
mutate(across(c(bf05_50_t1_r, bf05_55_t1_r, bf05_30_t1_r, bf05_25_t1_r, bf05_45_t1_r, bf05_05_t1_r,
bf05_50_t2_r, bf05_55_t2_r, bf05_30_t2_r, bf05_25_t2_r, bf05_45_t2_r, bf05_05_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(openn_curr_par1_t1 = rowMeans(across(c(bf05_30_t1_r, bf05_60_t1, bf05_15_t1, bf05_05_t1_r)), na.rm=T),
openn_curr_par2_t1 = rowMeans(across(c(bf05_35_t1, bf05_55_t1_r, bf05_25_t1_r, bf05_40_t1)), na.rm=T),
openn_curr_par3_t1 = rowMeans(across(c(bf05_50_t1_r, bf05_20_t1, bf05_45_t1_r, bf05_10_t1)), na.rm=T),
openn_curr_par1_t2 = rowMeans(across(c(bf05_30_t2_r, bf05_60_t2, bf05_15_t2, bf05_05_t2_r)), na.rm=T),
openn_curr_par2_t2 = rowMeans(across(c(bf05_35_t2, bf05_55_t2_r, bf05_25_t2_r, bf05_40_t2)), na.rm=T),
openn_curr_par3_t2 = rowMeans(across(c(bf05_50_t2_r, bf05_20_t2, bf05_45_t2_r, bf05_10_t2)), na.rm=T))Check CFA item loadings at T1
cfa_openn_ideal <- '
# Define the latent factors
openn_ideal1 =~ NA*bf06_05_t1 + lambda1*bf06_05_t1 + lambda2*bf06_10_t1 + lambda3*bf06_15_t1 + lambda4*bf06_20_t1 + lambda5*bf06_25_t1 + lambda6*bf06_30_t1 + lambda7*bf06_35_t1 + lambda8*bf06_40_t1 + lambda9*bf06_45_t1 + lambda10*bf06_50_t1 + lambda11*bf06_55_t1 + lambda12*bf06_60_t1
# Intercepts
bf06_05_t1 ~ i1*1
bf06_10_t1 ~ 1
bf06_15_t1 ~ 1
bf06_20_t1 ~ 1
bf06_25_t1 ~ 1
bf06_30_t1 ~ 1
bf06_35_t1 ~ 1
bf06_40_t1 ~ 1
bf06_45_t1 ~ 1
bf06_50_t1 ~ 1
bf06_55_t1 ~ 1
bf06_60_t1 ~ 1
# Unique Variances
bf06_05_t1 ~~ bf06_05_t1
bf06_10_t1 ~~ bf06_10_t1
bf06_15_t1 ~~ bf06_15_t1
bf06_20_t1 ~~ bf06_20_t1
bf06_25_t1 ~~ bf06_25_t1
bf06_30_t1 ~~ bf06_30_t1
bf06_35_t1 ~~ bf06_35_t1
bf06_40_t1 ~~ bf06_40_t1
bf06_45_t1 ~~ bf06_45_t1
bf06_50_t1 ~~ bf06_50_t1
bf06_55_t1 ~~ bf06_55_t1
bf06_60_t1 ~~ bf06_60_t1
# Latent Variable Means
openn_ideal1 ~ 0*1
# Latent Variable Variances and Covariance
openn_ideal1 ~~ 1*openn_ideal1
'
fit_cfa_openn_ideal <- cfa(cfa_openn_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_cfa_openn_ideal, fit.measures = TRUE)lavaan 0.6.15 ended normally after 43 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 36
Used Total
Number of observations 174 175
Number of missing patterns 2
Model Test User Model:
Test statistic 91.003
Degrees of freedom 54
P-value (Chi-square) 0.001
Model Test Baseline Model:
Test statistic 409.313
Degrees of freedom 66
P-value 0.000
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.892
Tucker-Lewis Index (TLI) 0.868
Robust Comparative Fit Index (CFI) 0.892
Robust Tucker-Lewis Index (TLI) 0.868
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -2848.673
Loglikelihood unrestricted model (H1) -2803.172
Akaike (AIC) 5769.347
Bayesian (BIC) 5883.073
Sample-size adjusted Bayesian (SABIC) 5769.074
Root Mean Square Error of Approximation:
RMSEA 0.063
90 Percent confidence interval - lower 0.039
90 Percent confidence interval - upper 0.085
P-value H_0: RMSEA <= 0.050 0.169
P-value H_0: RMSEA >= 0.080 0.102
Robust RMSEA 0.063
90 Percent confidence interval - lower 0.039
90 Percent confidence interval - upper 0.085
P-value H_0: Robust RMSEA <= 0.050 0.168
P-value H_0: Robust RMSEA >= 0.080 0.103
Standardized Root Mean Square Residual:
SRMR 0.057
Parameter Estimates:
Standard errors Standard
Information Observed
Observed information based on Hessian
Latent Variables:
Estimate Std.Err z-value P(>|z|)
openn_ideal1 =~
b06_05_ (lmb1) 0.140 0.126 1.112 0.266
b06_10_ (lmb2) -0.305 0.067 -4.559 0.000
b06_15_ (lmb3) -0.332 0.055 -6.002 0.000
b06_20_ (lmb4) -0.544 0.073 -7.405 0.000
b06_25_ (lmb5) 0.453 0.078 5.782 0.000
b06_30_ (lmb6) 0.662 0.109 6.048 0.000
b06_35_ (lmb7) -0.611 0.069 -8.887 0.000
b06_40_ (lmb8) -0.457 0.091 -5.033 0.000
b06_45_ (lmb9) 0.599 0.088 6.787 0.000
b06_50_ (lm10) 0.447 0.094 4.756 0.000
b06_55_ (lm11) 0.599 0.093 6.465 0.000
b06_60_ (lm12) -0.523 0.059 -8.796 0.000
Intercepts:
Estimate Std.Err z-value P(>|z|)
.bf06_05_1 (i1) 3.144 0.113 27.936 0.000
.bf06_10_1 4.454 0.062 71.608 0.000
.bf06_15_1 4.603 0.053 87.585 0.000
.bf06_20_1 4.305 0.072 60.189 0.000
.bf06_25_1 1.672 0.075 22.429 0.000
.bf06_30_1 2.029 0.104 19.437 0.000
.bf06_35_1 4.310 0.069 62.665 0.000
.bf06_40_1 3.931 0.085 46.202 0.000
.bf06_45_1 1.707 0.085 20.113 0.000
.bf06_50_1 2.011 0.088 22.980 0.000
.bf06_55_1 2.103 0.089 23.618 0.000
.bf06_60_1 4.489 0.060 75.334 0.000
openn_dl1 0.000
Variances:
Estimate Std.Err z-value P(>|z|)
.bf06_05_t1 2.184 0.235 9.307 0.000
.bf06_10_t1 0.580 0.065 8.963 0.000
.bf06_15_t1 0.370 0.043 8.617 0.000
.bf06_20_t1 0.594 0.073 8.086 0.000
.bf06_25_t1 0.762 0.087 8.725 0.000
.bf06_30_t1 1.449 0.169 8.584 0.000
.bf06_35_t1 0.450 0.063 7.201 0.000
.bf06_40_t1 1.051 0.118 8.870 0.000
.bf06_45_t1 0.895 0.107 8.340 0.000
.bf06_50_t1 1.133 0.127 8.913 0.000
.bf06_55_t1 1.022 0.120 8.514 0.000
.bf06_60_t1 0.344 0.047 7.316 0.000
openn_ideal1 1.000 tidy(fit_cfa_openn_ideal) %>% filter(str_detect(label, "lambda")) %>% mutate(abs_loading = abs(estimate)) %>% arrange(desc(abs_loading)) %>% select(1:5, std.all)# A tibble: 12 × 6
term op label estimate std.error std.all
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 openn_ideal1 =~ bf06_30_t1 =~ lambda6 0.662 0.109 0.482
2 openn_ideal1 =~ bf06_35_t1 =~ lambda7 -0.611 0.0687 -0.673
3 openn_ideal1 =~ bf06_45_t1 =~ lambda9 0.599 0.0882 0.535
4 openn_ideal1 =~ bf06_55_t1 =~ lambda11 0.599 0.0926 0.510
5 openn_ideal1 =~ bf06_20_t1 =~ lambda4 -0.544 0.0735 -0.577
6 openn_ideal1 =~ bf06_60_t1 =~ lambda12 -0.523 0.0595 -0.666
7 openn_ideal1 =~ bf06_40_t1 =~ lambda8 -0.457 0.0908 -0.407
8 openn_ideal1 =~ bf06_25_t1 =~ lambda5 0.453 0.0783 0.460
9 openn_ideal1 =~ bf06_50_t1 =~ lambda10 0.447 0.0940 0.387
10 openn_ideal1 =~ bf06_15_t1 =~ lambda3 -0.332 0.0553 -0.479
11 openn_ideal1 =~ bf06_10_t1 =~ lambda2 -0.305 0.0668 -0.371
12 openn_ideal1 =~ bf06_05_t1 =~ lambda1 0.140 0.126 0.0940Reverse-code and form parcels:
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(bf06_50_t1_r = bf06_50_t1,
bf06_55_t1_r = bf06_55_t1,
bf06_30_t1_r = bf06_30_t1,
bf06_25_t1_r = bf06_25_t1,
bf06_45_t1_r = bf06_45_t1,
bf06_05_t1_r = bf06_05_t1,
bf06_50_t2_r = bf06_50_t2,
bf06_55_t2_r = bf06_55_t2,
bf06_30_t2_r = bf06_30_t2,
bf06_25_t2_r = bf06_25_t2,
bf06_45_t2_r = bf06_45_t2,
bf06_05_t2_r = bf06_05_t2) %>%
mutate(across(c(bf06_50_t1_r, bf06_55_t1_r, bf06_30_t1_r, bf06_25_t1_r, bf06_45_t1_r, bf06_05_t1_r,
bf06_50_t2_r, bf06_55_t2_r, bf06_30_t2_r, bf06_25_t2_r, bf06_45_t2_r, bf06_05_t2_r),
~ recode(.x, `1` = 5L, `2` = 4L, `3` = 3L, `4` = 2L, `5` = 1L, .default = NA_integer_))) %>%
mutate(openn_ideal_par1_t1 = rowMeans(across(c(bf06_30_t1_r, bf06_60_t1, bf06_50_t1_r, bf06_05_t1_r)), na.rm=T),
openn_ideal_par2_t1 = rowMeans(across(c(bf06_35_t1, bf06_20_t1, bf06_25_t1_r, bf06_10_t1)), na.rm=T),
openn_ideal_par3_t1 = rowMeans(across(c(bf06_45_t1_r, bf06_55_t1_r, bf06_40_t1, bf06_15_t1)), na.rm=T),
openn_ideal_par1_t2 = rowMeans(across(c(bf06_30_t2_r, bf06_60_t2, bf06_50_t2_r, bf06_05_t2_r)), na.rm=T),
openn_ideal_par2_t2 = rowMeans(across(c(bf06_35_t2, bf06_20_t2, bf06_25_t2_r, bf06_10_t2)), na.rm=T),
openn_ideal_par3_t2 = rowMeans(across(c(bf06_45_t2_r, bf06_55_t2_r, bf06_40_t2, bf06_15_t2)), na.rm=T))
# replace NaN with regular NA
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate_all(~ifelse(is.nan(.), NA, .))This is only relevant for the facet-level models. All trait-level models rely on the parcels created above. However, the facet-level models pick the first item as the anchor indicator which is sometimes reverse-keyed for the latent factor. For analyses below I want to use just positively keyed variables (for neuroticism facets higher values should indicate higher neuroticism).
# save all relevant variable names
names_rev_scored <- df_sbsa3_wide_pers %>% select(ends_with("_r")) %>% colnames()
names_rev_scored_orig <- gsub("_r", "", names_rev_scored) # variable names without the "_r" -> these are the original variables
# drop original variables
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>% select(-all_of(names_rev_scored_orig))
# recreate original variable names -> these are the recoded variables now!
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>% rename_with(~ gsub("_r", "", .x))Testing for measurement invariance of the latent traits across time.
Satisfaction with life scale (only the first four items - last one taps into a somewhat different construct)
Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_swls <- '
# Define the latent factors
swls1 =~ NA*sw06_01_t1 + lambda1*sw06_01_t1 + sw06_02_t1 + sw06_03_t1 + sw06_04_t1
swls2 =~ NA*sw06_01_t2 + lambda1*sw06_01_t2 + sw06_02_t2 + sw06_03_t2 + sw06_04_t2
# Intercepts
sw06_01_t1 ~ i1*1
sw06_02_t1 ~ 1
sw06_03_t1 ~ 1
sw06_04_t1 ~ 1
sw06_01_t2 ~ i1*1
sw06_02_t2 ~ 1
sw06_03_t2 ~ 1
sw06_04_t2 ~ 1
# Unique Variances
sw06_01_t1 ~~ sw06_01_t1
sw06_02_t1 ~~ sw06_02_t1
sw06_03_t1 ~~ sw06_03_t1
sw06_04_t1 ~~ sw06_04_t1
sw06_01_t2 ~~ sw06_01_t2
sw06_02_t2 ~~ sw06_02_t2
sw06_03_t2 ~~ sw06_03_t2
sw06_04_t2 ~~ sw06_04_t2
# Latent Variable Means
swls1 ~ 0*1
swls2 ~ 1
# Latent Variable Variances and Covariance
swls1 ~~ 1*swls1
swls2 ~~ swls2
swls1 ~~ swls2
'
fit_configural_swls <- cfa(configural_swls, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_configural_swls, fit.measures = TRUE)
# Weak invariance model
weak_swls <- '
# Define the latent factors
swls1 =~ NA*sw06_01_t1 + lambda1*sw06_01_t1 + lambda2*sw06_02_t1 + lambda3*sw06_03_t1 + lambda4*sw06_04_t1
swls2 =~ NA*sw06_01_t2 + lambda1*sw06_01_t2 + lambda2*sw06_02_t2 + lambda3*sw06_03_t2 + lambda4*sw06_04_t2
# Intercepts
sw06_01_t1 ~ i1*1
sw06_02_t1 ~ 1
sw06_03_t1 ~ 1
sw06_04_t1 ~ 1
sw06_01_t2 ~ i1*1
sw06_02_t2 ~ 1
sw06_03_t2 ~ 1
sw06_04_t2 ~ 1
# Unique Variances
sw06_01_t1 ~~ sw06_01_t1
sw06_02_t1 ~~ sw06_02_t1
sw06_03_t1 ~~ sw06_03_t1
sw06_04_t1 ~~ sw06_04_t1
sw06_01_t2 ~~ sw06_01_t2
sw06_02_t2 ~~ sw06_02_t2
sw06_03_t2 ~~ sw06_03_t2
sw06_04_t2 ~~ sw06_04_t2
# Latent Variable Means
swls1 ~ 0*1
swls2 ~ 1
# Latent Variable Variances and Covariance
swls1 ~~ 1*swls1
swls2 ~~ swls2
swls1 ~~ swls2
'
fit_weak_swls <- cfa(weak_swls, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_weak_swls, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_swls <- '
# Define the latent factors
swls1 =~ NA*sw06_01_t1 + lambda1*sw06_01_t1 + lambda2*sw06_02_t1 + lambda3*sw06_03_t1 + lambda4*sw06_04_t1
swls2 =~ NA*sw06_01_t2 + lambda1*sw06_01_t2 + lambda2*sw06_02_t2 + lambda3*sw06_03_t2 + lambda4*sw06_04_t2
# Intercepts
sw06_01_t1 ~ i1*1
sw06_02_t1 ~ i2*1
sw06_03_t1 ~ i3*1
sw06_04_t1 ~ i4*1
sw06_01_t2 ~ i1*1
sw06_02_t2 ~ i2*1
sw06_03_t2 ~ i3*1
sw06_04_t2 ~ i4*1
# Unique Variances
sw06_01_t1 ~~ sw06_01_t1
sw06_02_t1 ~~ sw06_02_t1
sw06_03_t1 ~~ sw06_03_t1
sw06_04_t1 ~~ sw06_04_t1
sw06_01_t2 ~~ sw06_01_t2
sw06_02_t2 ~~ sw06_02_t2
sw06_03_t2 ~~ sw06_03_t2
sw06_04_t2 ~~ sw06_04_t2
# Latent Variable Means
swls1 ~ 0*1
swls2 ~ 1
# Latent Variable Variances and Covariance
swls1 ~~ 1*swls1
swls2 ~~ swls2
swls1 ~~ swls2
'
fit_strong_swls <- cfa(strong_swls, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_strong_swls, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_swls <- '
# Define the latent factors
swls1 =~ NA*sw06_01_t1 + lambda1*sw06_01_t1 + lambda2*sw06_02_t1 + lambda3*sw06_03_t1 + lambda4*sw06_04_t1
swls2 =~ NA*sw06_01_t2 + lambda1*sw06_01_t2 + lambda2*sw06_02_t2 + lambda3*sw06_03_t2 + lambda4*sw06_04_t2
# Intercepts
sw06_01_t1 ~ i1*1
sw06_02_t1 ~ i2*1
sw06_03_t1 ~ i3*1
sw06_04_t1 ~ i4*1
sw06_01_t2 ~ i1*1
sw06_02_t2 ~ i2*1
sw06_03_t2 ~ i3*1
sw06_04_t2 ~ i4*1
# Unique Variances
sw06_01_t1 ~~ u1*sw06_01_t1
sw06_02_t1 ~~ u2*sw06_02_t1
sw06_03_t1 ~~ u3*sw06_03_t1
sw06_04_t1 ~~ u4*sw06_04_t1
sw06_01_t2 ~~ u1*sw06_01_t2
sw06_02_t2 ~~ u2*sw06_02_t2
sw06_03_t2 ~~ u3*sw06_03_t2
sw06_04_t2 ~~ u4*sw06_04_t2
# Latent Variable Means
swls1 ~ 0*1
swls2 ~ 1
# Latent Variable Variances and Covariance
swls1 ~~ 1*swls1
swls2 ~~ swls2
swls1 ~~ swls2
'
fit_strict_swls <- cfa(strict_swls, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_strict_swls, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_swls) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_swls) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_swls) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_swls) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 25 241. 10427. 10534. 0.917 0.877 0.148 0.0442
2 weak 530 22 244. 10425. 10519. 0.917 0.894 0.138 0.0491
3 strong 530 19 244. 10419. 10500. 0.918 0.908 0.129 0.0491
4 strict 530 15 250. 10417. 10481. 0.917 0.920 0.120 0.0485# chi-square difference test for nested models
anova(fit_configural_swls, fit_weak_swls)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_swls 19 10427 10534 240.57
fit_weak_swls 22 10424 10518 243.84 3.2664 0.012944 3
Pr(>Chisq)
fit_configural_swls
fit_weak_swls 0.3523anova(fit_weak_swls, fit_strong_swls)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
fit_weak_swls 22 10424 10518 243.84
fit_strong_swls 25 10419 10500 244.08 0.24147 0 3 0.9706anova(fit_strong_swls, fit_strict_swls)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
fit_strong_swls 25 10419 10500 244.08
fit_strict_swls 29 10416 10481 249.85 5.771 0.028903 4 0.2169Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_meaning <- '
# Define the latent factors
meaning1 =~ NA*meaning_par1_t1 + lambda1*meaning_par1_t1 + meaning_par2_t1 + meaning_par3_t1
meaning2 =~ NA*meaning_par1_t2 + lambda1*meaning_par1_t2 + meaning_par2_t2 + meaning_par3_t2
# Intercepts
meaning_par1_t1 ~ i1*1
meaning_par2_t1 ~ 1
meaning_par3_t1 ~ 1
meaning_par1_t2 ~ i1*1
meaning_par2_t2 ~ 1
meaning_par3_t2 ~ 1
# Unique Variances
meaning_par1_t1 ~~ meaning_par1_t1
meaning_par2_t1 ~~ meaning_par2_t1
meaning_par3_t1 ~~ meaning_par3_t1
meaning_par1_t2 ~~ meaning_par1_t2
meaning_par2_t2 ~~ meaning_par2_t2
meaning_par3_t2 ~~ meaning_par3_t2
# Latent Variable Means
meaning1 ~ 0*1
meaning2 ~ 1
# Latent Variable Variances and Covariance
meaning1 ~~ 1*meaning1
meaning2 ~~ meaning2
meaning1 ~~ meaning2
'
fit_configural_meaning <- cfa(configural_meaning, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_configural_meaning, fit.measures = TRUE)
# Weak invariance model
weak_meaning <- '
# Define the latent factors
meaning1 =~ NA*meaning_par1_t1 + lambda1*meaning_par1_t1 + lambda2*meaning_par2_t1 + lambda3*meaning_par3_t1
meaning2 =~ NA*meaning_par1_t2 + lambda1*meaning_par1_t2 + lambda2*meaning_par2_t2 + lambda3*meaning_par3_t2
# Intercepts
meaning_par1_t1 ~ i1*1
meaning_par2_t1 ~ 1
meaning_par3_t1 ~ 1
meaning_par1_t2 ~ i1*1
meaning_par2_t2 ~ 1
meaning_par3_t2 ~ 1
# Unique Variances
meaning_par1_t1 ~~ meaning_par1_t1
meaning_par2_t1 ~~ meaning_par2_t1
meaning_par3_t1 ~~ meaning_par3_t1
meaning_par1_t2 ~~ meaning_par1_t2
meaning_par2_t2 ~~ meaning_par2_t2
meaning_par3_t2 ~~ meaning_par3_t2
# Latent Variable Means
meaning1 ~ 0*1
meaning2 ~ 1
# Latent Variable Variances and Covariance
meaning1 ~~ 1*meaning1
meaning2 ~~ meaning2
meaning1 ~~ meaning2
'
fit_weak_meaning <- cfa(weak_meaning, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_weak_meaning, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_meaning <- '
# Define the latent factors
meaning1 =~ NA*meaning_par1_t1 + lambda1*meaning_par1_t1 + lambda2*meaning_par2_t1 + lambda3*meaning_par3_t1
meaning2 =~ NA*meaning_par1_t2 + lambda1*meaning_par1_t2 + lambda2*meaning_par2_t2 + lambda3*meaning_par3_t2
# Intercepts
meaning_par1_t1 ~ i1*1
meaning_par2_t1 ~ i2*1
meaning_par3_t1 ~ i3*1
meaning_par1_t2 ~ i1*1
meaning_par2_t2 ~ i2*1
meaning_par3_t2 ~ i3*1
# Unique Variances
meaning_par1_t1 ~~ meaning_par1_t1
meaning_par2_t1 ~~ meaning_par2_t1
meaning_par3_t1 ~~ meaning_par3_t1
meaning_par1_t2 ~~ meaning_par1_t2
meaning_par2_t2 ~~ meaning_par2_t2
meaning_par3_t2 ~~ meaning_par3_t2
# Latent Variable Means
meaning1 ~ 0*1
meaning2 ~ 1
# Latent Variable Variances and Covariance
meaning1 ~~ 1*meaning1
meaning2 ~~ meaning2
meaning1 ~~ meaning2
'
fit_strong_meaning <- cfa(strong_meaning, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_strong_meaning, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_meaning <- '
# Define the latent factors
meaning1 =~ NA*meaning_par1_t1 + lambda1*meaning_par1_t1 + lambda2*meaning_par2_t1 + lambda3*meaning_par3_t1
meaning2 =~ NA*meaning_par1_t2 + lambda1*meaning_par1_t2 + lambda2*meaning_par2_t2 + lambda3*meaning_par3_t2
# Intercepts
meaning_par1_t1 ~ i1*1
meaning_par2_t1 ~ i2*1
meaning_par3_t1 ~ i3*1
meaning_par1_t2 ~ i1*1
meaning_par2_t2 ~ i2*1
meaning_par3_t2 ~ i3*1
# Unique Variances
meaning_par1_t1 ~~ u1*meaning_par1_t1
meaning_par2_t1 ~~ u2*meaning_par2_t1
meaning_par3_t1 ~~ u3*meaning_par3_t1
meaning_par1_t2 ~~ u1*meaning_par1_t2
meaning_par2_t2 ~~ u2*meaning_par2_t2
meaning_par3_t2 ~~ u3*meaning_par3_t2
# Latent Variable Means
meaning1 ~ 0*1
meaning2 ~ 1
# Latent Variable Variances and Covariance
meaning1 ~~ 1*meaning1
meaning2 ~~ meaning2
meaning1 ~~ meaning2
'
fit_strict_meaning <- cfa(strict_meaning, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_strict_meaning, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_meaning) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_meaning) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_meaning) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_meaning) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 19 193. 7726. 7807. 0.922 0.854 0.209 0.0432
2 weak 530 17 193. 7722. 7795. 0.923 0.884 0.186 0.0436
3 strong 530 15 195. 7720. 7784. 0.923 0.904 0.169 0.0447
4 strict 530 12 197. 7716. 7768. 0.923 0.923 0.151 0.0440# chi-square difference test for nested models
anova(fit_configural_meaning, fit_weak_meaning)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_meaning 8 7726.2 7807.4 192.68
fit_weak_meaning 10 7722.3 7795.0 192.86 0.17863 0 2
Pr(>Chisq)
fit_configural_meaning
fit_weak_meaning 0.9146anova(fit_weak_meaning, fit_strong_meaning)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
fit_weak_meaning 10 7722.3 7795.0 192.86
fit_strong_meaning 12 7720.1 7784.2 194.65 1.7902 0 2 0.4086anova(fit_strong_meaning, fit_strict_meaning)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
fit_strong_meaning 12 7720.1 7784.2 194.65
fit_strict_meaning 15 7716.4 7767.7 196.91 2.264 0 3 0.5195Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_selfes <- '
# Define the latent factors
selfes1 =~ NA*selfes_par1_t1 + lambda1*selfes_par1_t1 + selfes_par2_t1 + selfes_par3_t1
selfes2 =~ NA*selfes_par1_t2 + lambda1*selfes_par1_t2 + selfes_par2_t2 + selfes_par3_t2
# Intercepts
selfes_par1_t1 ~ i1*1
selfes_par2_t1 ~ 1
selfes_par3_t1 ~ 1
selfes_par1_t2 ~ i1*1
selfes_par2_t2 ~ 1
selfes_par3_t2 ~ 1
# Unique Variances
selfes_par1_t1 ~~ selfes_par1_t1
selfes_par2_t1 ~~ selfes_par2_t1
selfes_par3_t1 ~~ selfes_par3_t1
selfes_par1_t2 ~~ selfes_par1_t2
selfes_par2_t2 ~~ selfes_par2_t2
selfes_par3_t2 ~~ selfes_par3_t2
# Latent Variable Means
selfes1 ~ 0*1
selfes2 ~ 1
# Latent Variable Variances and Covariance
selfes1 ~~ 1*selfes1
selfes2 ~~ selfes2
selfes1 ~~ selfes2
'
fit_configural_selfes <- cfa(configural_selfes, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_configural_selfes, fit.measures = TRUE)
# Weak invariance model
weak_selfes <- '
# Define the latent factors
selfes1 =~ NA*selfes_par1_t1 + lambda1*selfes_par1_t1 + lambda2*selfes_par2_t1 + lambda3*selfes_par3_t1
selfes2 =~ NA*selfes_par1_t2 + lambda1*selfes_par1_t2 + lambda2*selfes_par2_t2 + lambda3*selfes_par3_t2
# Intercepts
selfes_par1_t1 ~ i1*1
selfes_par2_t1 ~ 1
selfes_par3_t1 ~ 1
selfes_par1_t2 ~ i1*1
selfes_par2_t2 ~ 1
selfes_par3_t2 ~ 1
# Unique Variances
selfes_par1_t1 ~~ selfes_par1_t1
selfes_par2_t1 ~~ selfes_par2_t1
selfes_par3_t1 ~~ selfes_par3_t1
selfes_par1_t2 ~~ selfes_par1_t2
selfes_par2_t2 ~~ selfes_par2_t2
selfes_par3_t2 ~~ selfes_par3_t2
# Latent Variable Means
selfes1 ~ 0*1
selfes2 ~ 1
# Latent Variable Variances and Covariance
selfes1 ~~ 1*selfes1
selfes2 ~~ selfes2
selfes1 ~~ selfes2
'
fit_weak_selfes <- cfa(weak_selfes, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_weak_selfes, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_selfes <- '
# Define the latent factors
selfes1 =~ NA*selfes_par1_t1 + lambda1*selfes_par1_t1 + lambda2*selfes_par2_t1 + lambda3*selfes_par3_t1
selfes2 =~ NA*selfes_par1_t2 + lambda1*selfes_par1_t2 + lambda2*selfes_par2_t2 + lambda3*selfes_par3_t2
# Intercepts
selfes_par1_t1 ~ i1*1
selfes_par2_t1 ~ i2*1
selfes_par3_t1 ~ i3*1
selfes_par1_t2 ~ i1*1
selfes_par2_t2 ~ i2*1
selfes_par3_t2 ~ i3*1
# Unique Variances
selfes_par1_t1 ~~ selfes_par1_t1
selfes_par2_t1 ~~ selfes_par2_t1
selfes_par3_t1 ~~ selfes_par3_t1
selfes_par1_t2 ~~ selfes_par1_t2
selfes_par2_t2 ~~ selfes_par2_t2
selfes_par3_t2 ~~ selfes_par3_t2
# Latent Variable Means
selfes1 ~ 0*1
selfes2 ~ 1
# Latent Variable Variances and Covariance
selfes1 ~~ 1*selfes1
selfes2 ~~ selfes2
selfes1 ~~ selfes2
'
fit_strong_selfes <- cfa(strong_selfes, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_strong_selfes, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_selfes <- '
# Define the latent factors
selfes1 =~ NA*selfes_par1_t1 + lambda1*selfes_par1_t1 + lambda2*selfes_par2_t1 + lambda3*selfes_par3_t1
selfes2 =~ NA*selfes_par1_t2 + lambda1*selfes_par1_t2 + lambda2*selfes_par2_t2 + lambda3*selfes_par3_t2
# Intercepts
selfes_par1_t1 ~ i1*1
selfes_par2_t1 ~ i2*1
selfes_par3_t1 ~ i3*1
selfes_par1_t2 ~ i1*1
selfes_par2_t2 ~ i2*1
selfes_par3_t2 ~ i3*1
# Unique Variances
selfes_par1_t1 ~~ u1*selfes_par1_t1
selfes_par2_t1 ~~ u2*selfes_par2_t1
selfes_par3_t1 ~~ u3*selfes_par3_t1
selfes_par1_t2 ~~ u1*selfes_par1_t2
selfes_par2_t2 ~~ u2*selfes_par2_t2
selfes_par3_t2 ~~ u3*selfes_par3_t2
# Latent Variable Means
selfes1 ~ 0*1
selfes2 ~ 1
# Latent Variable Variances and Covariance
selfes1 ~~ 1*selfes1
selfes2 ~~ selfes2
selfes1 ~~ selfes2
'
fit_strict_selfes <- cfa(strict_selfes, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_strict_selfes, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_selfes) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_selfes) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_selfes) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_selfes) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 19 164. 5667. 5748. 0.945 0.897 0.192 0.0275
2 weak 530 17 164. 5663. 5736. 0.946 0.919 0.170 0.0278
3 strong 530 15 164. 5660. 5724. 0.946 0.933 0.155 0.0282
4 strict 530 12 167. 5656. 5708. 0.946 0.946 0.138 0.0311# chi-square difference test for nested models
anova(fit_configural_selfes, fit_weak_selfes)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_selfes 8 5667.1 5748.2 163.71
fit_weak_selfes 10 5663.1 5735.8 163.79 0.081813 0 2
Pr(>Chisq)
fit_configural_selfes
fit_weak_selfes 0.9599anova(fit_weak_selfes, fit_strong_selfes)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
fit_weak_selfes 10 5663.1 5735.8 163.79
fit_strong_selfes 12 5659.5 5723.6 164.20 0.40974 0 2 0.8148anova(fit_strong_selfes, fit_strict_selfes)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
fit_strong_selfes 12 5659.5 5723.6 164.20
fit_strict_selfes 15 5656.3 5707.6 166.98 2.7759 0 3 0.4275Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_concept <- '
# Define the latent factors
concept1 =~ NA*concept_par1_t1 + lambda1*concept_par1_t1 + concept_par2_t1 + concept_par3_t1
concept2 =~ NA*concept_par1_t2 + lambda1*concept_par1_t2 + concept_par2_t2 + concept_par3_t2
# Intercepts
concept_par1_t1 ~ i1*1
concept_par2_t1 ~ 1
concept_par3_t1 ~ 1
concept_par1_t2 ~ i1*1
concept_par2_t2 ~ 1
concept_par3_t2 ~ 1
# Unique Variances
concept_par1_t1 ~~ concept_par1_t1
concept_par2_t1 ~~ concept_par2_t1
concept_par3_t1 ~~ concept_par3_t1
concept_par1_t2 ~~ concept_par1_t2
concept_par2_t2 ~~ concept_par2_t2
concept_par3_t2 ~~ concept_par3_t2
# Latent Variable Means
concept1 ~ 0*1
concept2 ~ 1
# Latent Variable Variances and Covariance
concept1 ~~ 1*concept1
concept2 ~~ concept2
concept1 ~~ concept2
'
fit_configural_concept <- cfa(configural_concept, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_configural_concept, fit.measures = TRUE)
# Weak invariance model
weak_concept <- '
# Define the latent factors
concept1 =~ NA*concept_par1_t1 + lambda1*concept_par1_t1 + lambda2*concept_par2_t1 + lambda3*concept_par3_t1
concept2 =~ NA*concept_par1_t2 + lambda1*concept_par1_t2 + lambda2*concept_par2_t2 + lambda3*concept_par3_t2
# Intercepts
concept_par1_t1 ~ i1*1
concept_par2_t1 ~ 1
concept_par3_t1 ~ 1
concept_par1_t2 ~ i1*1
concept_par2_t2 ~ 1
concept_par3_t2 ~ 1
# Unique Variances
concept_par1_t1 ~~ concept_par1_t1
concept_par2_t1 ~~ concept_par2_t1
concept_par3_t1 ~~ concept_par3_t1
concept_par1_t2 ~~ concept_par1_t2
concept_par2_t2 ~~ concept_par2_t2
concept_par3_t2 ~~ concept_par3_t2
# Latent Variable Means
concept1 ~ 0*1
concept2 ~ 1
# Latent Variable Variances and Covariance
concept1 ~~ 1*concept1
concept2 ~~ concept2
concept1 ~~ concept2
'
fit_weak_concept <- cfa(weak_concept, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_weak_concept, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_concept <- '
# Define the latent factors
concept1 =~ NA*concept_par1_t1 + lambda1*concept_par1_t1 + lambda2*concept_par2_t1 + lambda3*concept_par3_t1
concept2 =~ NA*concept_par1_t2 + lambda1*concept_par1_t2 + lambda2*concept_par2_t2 + lambda3*concept_par3_t2
# Intercepts
concept_par1_t1 ~ i1*1
concept_par2_t1 ~ i2*1
concept_par3_t1 ~ i3*1
concept_par1_t2 ~ i1*1
concept_par2_t2 ~ i2*1
concept_par3_t2 ~ i3*1
# Unique Variances
concept_par1_t1 ~~ concept_par1_t1
concept_par2_t1 ~~ concept_par2_t1
concept_par3_t1 ~~ concept_par3_t1
concept_par1_t2 ~~ concept_par1_t2
concept_par2_t2 ~~ concept_par2_t2
concept_par3_t2 ~~ concept_par3_t2
# Latent Variable Means
concept1 ~ 0*1
concept2 ~ 1
# Latent Variable Variances and Covariance
concept1 ~~ 1*concept1
concept2 ~~ concept2
concept1 ~~ concept2
'
fit_strong_concept <- cfa(strong_concept, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_strong_concept, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_concept <- '
# Define the latent factors
concept1 =~ NA*concept_par1_t1 + lambda1*concept_par1_t1 + lambda2*concept_par2_t1 + lambda3*concept_par3_t1
concept2 =~ NA*concept_par1_t2 + lambda1*concept_par1_t2 + lambda2*concept_par2_t2 + lambda3*concept_par3_t2
# Intercepts
concept_par1_t1 ~ i1*1
concept_par2_t1 ~ i2*1
concept_par3_t1 ~ i3*1
concept_par1_t2 ~ i1*1
concept_par2_t2 ~ i2*1
concept_par3_t2 ~ i3*1
# Unique Variances
concept_par1_t1 ~~ u1*concept_par1_t1
concept_par2_t1 ~~ u2*concept_par2_t1
concept_par3_t1 ~~ u3*concept_par3_t1
concept_par1_t2 ~~ u1*concept_par1_t2
concept_par2_t2 ~~ u2*concept_par2_t2
concept_par3_t2 ~~ u3*concept_par3_t2
# Latent Variable Means
concept1 ~ 0*1
concept2 ~ 1
# Latent Variable Variances and Covariance
concept1 ~~ 1*concept1
concept2 ~~ concept2
concept1 ~~ concept2
'
fit_strict_concept <- cfa(strict_concept, data = df_sbsa3_wide_wb, mimic = "mplus", missing="ML")
summary(fit_strict_concept, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_concept) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_concept) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_concept) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_concept) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 19 139. 5953. 6034. 0.948 0.903 0.176 0.0253
2 weak 530 17 140. 5950. 6023. 0.949 0.923 0.157 0.0290
3 strong 530 15 140. 5946. 6010. 0.949 0.937 0.142 0.0294
4 strict 530 12 143. 5944. 5995. 0.949 0.949 0.127 0.0394# chi-square difference test for nested models
anova(fit_configural_concept, fit_weak_concept)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_concept 8 5953.2 6034.4 138.71
fit_weak_concept 10 5950.3 6023.0 139.81 1.1045 0 2
Pr(>Chisq)
fit_configural_concept
fit_weak_concept 0.5756anova(fit_weak_concept, fit_strong_concept)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
fit_weak_concept 10 5950.3 6023.0 139.81
fit_strong_concept 12 5946.4 6010.5 139.86 0.052991 0 2 0.9739anova(fit_strong_concept, fit_strict_concept)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_concept 12 5946.4 6010.5 139.86
fit_strict_concept 15 5943.8 5995.1 143.27 3.4048 0.015956 3
Pr(>Chisq)
fit_strong_concept
fit_strict_concept 0.3333Testing for measurement invariance of the latent BFI-2 traits across time.
Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_extra_curr <- '
# Define the latent factors
extra_curr1 =~ NA*extra_curr_par1_t1 + lambda1*extra_curr_par1_t1 + extra_curr_par2_t1 + extra_curr_par3_t1
extra_curr2 =~ NA*extra_curr_par1_t2 + lambda1*extra_curr_par1_t2 + extra_curr_par2_t2 + extra_curr_par3_t2
# Intercepts
extra_curr_par1_t1 ~ i1*1
extra_curr_par2_t1 ~ 1
extra_curr_par3_t1 ~ 1
extra_curr_par1_t2 ~ i1*1
extra_curr_par2_t2 ~ 1
extra_curr_par3_t2 ~ 1
# Unique Variances
extra_curr_par1_t1 ~~ extra_curr_par1_t1
extra_curr_par2_t1 ~~ extra_curr_par2_t1
extra_curr_par3_t1 ~~ extra_curr_par3_t1
extra_curr_par1_t2 ~~ extra_curr_par1_t2
extra_curr_par2_t2 ~~ extra_curr_par2_t2
extra_curr_par3_t2 ~~ extra_curr_par3_t2
# Latent Variable Means
extra_curr1 ~ 0*1
extra_curr2 ~ 1
# Latent Variable Variances and Covariance
extra_curr1 ~~ 1*extra_curr1
extra_curr2 ~~ extra_curr2
extra_curr1 ~~ extra_curr2
'
fit_configural_extra_curr <- cfa(configural_extra_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_configural_extra_curr, fit.measures = TRUE)
# Weak invariance model
weak_extra_curr <- '
# Define the latent factors
extra_curr1 =~ NA*extra_curr_par1_t1 + lambda1*extra_curr_par1_t1 + lambda2*extra_curr_par2_t1 + lambda3*extra_curr_par3_t1
extra_curr2 =~ NA*extra_curr_par1_t2 + lambda1*extra_curr_par1_t2 + lambda2*extra_curr_par2_t2 + lambda3*extra_curr_par3_t2
# Intercepts
extra_curr_par1_t1 ~ i1*1
extra_curr_par2_t1 ~ 1
extra_curr_par3_t1 ~ 1
extra_curr_par1_t2 ~ i1*1
extra_curr_par2_t2 ~ 1
extra_curr_par3_t2 ~ 1
# Unique Variances
extra_curr_par1_t1 ~~ extra_curr_par1_t1
extra_curr_par2_t1 ~~ extra_curr_par2_t1
extra_curr_par3_t1 ~~ extra_curr_par3_t1
extra_curr_par1_t2 ~~ extra_curr_par1_t2
extra_curr_par2_t2 ~~ extra_curr_par2_t2
extra_curr_par3_t2 ~~ extra_curr_par3_t2
# Latent Variable Means
extra_curr1 ~ 0*1
extra_curr2 ~ 1
# Latent Variable Variances and Covariance
extra_curr1 ~~ 1*extra_curr1
extra_curr2 ~~ extra_curr2
extra_curr1 ~~ extra_curr2
'
fit_weak_extra_curr <- cfa(weak_extra_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_weak_extra_curr, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_extra_curr <- '
# Define the latent factors
extra_curr1 =~ NA*extra_curr_par1_t1 + lambda1*extra_curr_par1_t1 + lambda2*extra_curr_par2_t1 + lambda3*extra_curr_par3_t1
extra_curr2 =~ NA*extra_curr_par1_t2 + lambda1*extra_curr_par1_t2 + lambda2*extra_curr_par2_t2 + lambda3*extra_curr_par3_t2
# Intercepts
extra_curr_par1_t1 ~ i1*1
extra_curr_par2_t1 ~ i2*1
extra_curr_par3_t1 ~ i3*1
extra_curr_par1_t2 ~ i1*1
extra_curr_par2_t2 ~ i2*1
extra_curr_par3_t2 ~ i3*1
# Unique Variances
extra_curr_par1_t1 ~~ extra_curr_par1_t1
extra_curr_par2_t1 ~~ extra_curr_par2_t1
extra_curr_par3_t1 ~~ extra_curr_par3_t1
extra_curr_par1_t2 ~~ extra_curr_par1_t2
extra_curr_par2_t2 ~~ extra_curr_par2_t2
extra_curr_par3_t2 ~~ extra_curr_par3_t2
# Latent Variable Means
extra_curr1 ~ 0*1
extra_curr2 ~ 1
# Latent Variable Variances and Covariance
extra_curr1 ~~ 1*extra_curr1
extra_curr2 ~~ extra_curr2
extra_curr1 ~~ extra_curr2
'
fit_strong_extra_curr <- cfa(strong_extra_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strong_extra_curr, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_extra_curr <- '
# Define the latent factors
extra_curr1 =~ NA*extra_curr_par1_t1 + lambda1*extra_curr_par1_t1 + lambda2*extra_curr_par2_t1 + lambda3*extra_curr_par3_t1
extra_curr2 =~ NA*extra_curr_par1_t2 + lambda1*extra_curr_par1_t2 + lambda2*extra_curr_par2_t2 + lambda3*extra_curr_par3_t2
# Intercepts
extra_curr_par1_t1 ~ i1*1
extra_curr_par2_t1 ~ i2*1
extra_curr_par3_t1 ~ i3*1
extra_curr_par1_t2 ~ i1*1
extra_curr_par2_t2 ~ i2*1
extra_curr_par3_t2 ~ i3*1
# Unique Variances
extra_curr_par1_t1 ~~ u1*extra_curr_par1_t1
extra_curr_par2_t1 ~~ u2*extra_curr_par2_t1
extra_curr_par3_t1 ~~ u3*extra_curr_par3_t1
extra_curr_par1_t2 ~~ u1*extra_curr_par1_t2
extra_curr_par2_t2 ~~ u2*extra_curr_par2_t2
extra_curr_par3_t2 ~~ u3*extra_curr_par3_t2
# Latent Variable Means
extra_curr1 ~ 0*1
extra_curr2 ~ 1
# Latent Variable Variances and Covariance
extra_curr1 ~~ 1*extra_curr1
extra_curr2 ~~ extra_curr2
extra_curr1 ~~ extra_curr2
'
fit_strict_extra_curr <- cfa(strict_extra_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strict_extra_curr, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_extra_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_extra_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_extra_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_extra_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 19 216. 5133. 5214. 0.919 0.848 0.222 0.0350
2 weak 530 17 218. 5130. 5203. 0.919 0.879 0.198 0.0390
3 strong 530 15 218. 5127. 5191. 0.920 0.899 0.180 0.0385
4 strict 530 12 221. 5124. 5175. 0.920 0.920 0.161 0.0409# chi-square difference test for nested models
anova(fit_configural_extra_curr, fit_weak_extra_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_extra_curr 8 5132.8 5214.0 216.27
fit_weak_extra_curr 10 5130.1 5202.7 217.54 1.2659 0 2
Pr(>Chisq)
fit_configural_extra_curr
fit_weak_extra_curr 0.531anova(fit_weak_extra_curr, fit_strong_extra_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_extra_curr 10 5130.1 5202.7 217.54
fit_strong_extra_curr 12 5127.0 5191.1 218.47 0.92766 0 2
Pr(>Chisq)
fit_weak_extra_curr
fit_strong_extra_curr 0.6289anova(fit_strong_extra_curr, fit_strict_extra_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_extra_curr 12 5127.0 5191.1 218.47
fit_strict_extra_curr 15 5123.8 5175.1 221.23 2.7637 0 3
Pr(>Chisq)
fit_strong_extra_curr
fit_strict_extra_curr 0.4295Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_extra_ideal <- '
# Define the latent factors
extra_ideal1 =~ NA*extra_ideal_par1_t1 + lambda1*extra_ideal_par1_t1 + extra_ideal_par2_t1 + extra_ideal_par3_t1
extra_ideal2 =~ NA*extra_ideal_par1_t2 + lambda1*extra_ideal_par1_t2 + extra_ideal_par2_t2 + extra_ideal_par3_t2
# Intercepts
extra_ideal_par1_t1 ~ i1*1
extra_ideal_par2_t1 ~ 1
extra_ideal_par3_t1 ~ 1
extra_ideal_par1_t2 ~ i1*1
extra_ideal_par2_t2 ~ 1
extra_ideal_par3_t2 ~ 1
# Unique Variances
extra_ideal_par1_t1 ~~ extra_ideal_par1_t1
extra_ideal_par2_t1 ~~ extra_ideal_par2_t1
extra_ideal_par3_t1 ~~ extra_ideal_par3_t1
extra_ideal_par1_t2 ~~ extra_ideal_par1_t2
extra_ideal_par2_t2 ~~ extra_ideal_par2_t2
extra_ideal_par3_t2 ~~ extra_ideal_par3_t2
# Latent Variable Means
extra_ideal1 ~ 0*1
extra_ideal2 ~ 1
# Latent Variable Variances and Covariance
extra_ideal1 ~~ 1*extra_ideal1
extra_ideal2 ~~ extra_ideal2
extra_ideal1 ~~ extra_ideal2
'
fit_configural_extra_ideal <- cfa(configural_extra_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_configural_extra_ideal, fit.measures = TRUE)
# Weak invariance model
weak_extra_ideal <- '
# Define the latent factors
extra_ideal1 =~ NA*extra_ideal_par1_t1 + lambda1*extra_ideal_par1_t1 + lambda2*extra_ideal_par2_t1 + lambda3*extra_ideal_par3_t1
extra_ideal2 =~ NA*extra_ideal_par1_t2 + lambda1*extra_ideal_par1_t2 + lambda2*extra_ideal_par2_t2 + lambda3*extra_ideal_par3_t2
# Intercepts
extra_ideal_par1_t1 ~ i1*1
extra_ideal_par2_t1 ~ 1
extra_ideal_par3_t1 ~ 1
extra_ideal_par1_t2 ~ i1*1
extra_ideal_par2_t2 ~ 1
extra_ideal_par3_t2 ~ 1
# Unique Variances
extra_ideal_par1_t1 ~~ extra_ideal_par1_t1
extra_ideal_par2_t1 ~~ extra_ideal_par2_t1
extra_ideal_par3_t1 ~~ extra_ideal_par3_t1
extra_ideal_par1_t2 ~~ extra_ideal_par1_t2
extra_ideal_par2_t2 ~~ extra_ideal_par2_t2
extra_ideal_par3_t2 ~~ extra_ideal_par3_t2
# Latent Variable Means
extra_ideal1 ~ 0*1
extra_ideal2 ~ 1
# Latent Variable Variances and Covariance
extra_ideal1 ~~ 1*extra_ideal1
extra_ideal2 ~~ extra_ideal2
extra_ideal1 ~~ extra_ideal2
'
fit_weak_extra_ideal <- cfa(weak_extra_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_weak_extra_ideal, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_extra_ideal <- '
# Define the latent factors
extra_ideal1 =~ NA*extra_ideal_par1_t1 + lambda1*extra_ideal_par1_t1 + lambda2*extra_ideal_par2_t1 + lambda3*extra_ideal_par3_t1
extra_ideal2 =~ NA*extra_ideal_par1_t2 + lambda1*extra_ideal_par1_t2 + lambda2*extra_ideal_par2_t2 + lambda3*extra_ideal_par3_t2
# Intercepts
extra_ideal_par1_t1 ~ i1*1
extra_ideal_par2_t1 ~ i2*1
extra_ideal_par3_t1 ~ i3*1
extra_ideal_par1_t2 ~ i1*1
extra_ideal_par2_t2 ~ i2*1
extra_ideal_par3_t2 ~ i3*1
# Unique Variances
extra_ideal_par1_t1 ~~ extra_ideal_par1_t1
extra_ideal_par2_t1 ~~ extra_ideal_par2_t1
extra_ideal_par3_t1 ~~ extra_ideal_par3_t1
extra_ideal_par1_t2 ~~ extra_ideal_par1_t2
extra_ideal_par2_t2 ~~ extra_ideal_par2_t2
extra_ideal_par3_t2 ~~ extra_ideal_par3_t2
# Latent Variable Means
extra_ideal1 ~ 0*1
extra_ideal2 ~ 1
# Latent Variable Variances and Covariance
extra_ideal1 ~~ 1*extra_ideal1
extra_ideal2 ~~ extra_ideal2
extra_ideal1 ~~ extra_ideal2
'
fit_strong_extra_ideal <- cfa(strong_extra_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_strong_extra_ideal, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_extra_ideal <- '
# Define the latent factors
extra_ideal1 =~ NA*extra_ideal_par1_t1 + lambda1*extra_ideal_par1_t1 + lambda2*extra_ideal_par2_t1 + lambda3*extra_ideal_par3_t1
extra_ideal2 =~ NA*extra_ideal_par1_t2 + lambda1*extra_ideal_par1_t2 + lambda2*extra_ideal_par2_t2 + lambda3*extra_ideal_par3_t2
# Intercepts
extra_ideal_par1_t1 ~ i1*1
extra_ideal_par2_t1 ~ i2*1
extra_ideal_par3_t1 ~ i3*1
extra_ideal_par1_t2 ~ i1*1
extra_ideal_par2_t2 ~ i2*1
extra_ideal_par3_t2 ~ i3*1
# Unique Variances
extra_ideal_par1_t1 ~~ u1*extra_ideal_par1_t1
extra_ideal_par2_t1 ~~ u2*extra_ideal_par2_t1
extra_ideal_par3_t1 ~~ u3*extra_ideal_par3_t1
extra_ideal_par1_t2 ~~ u1*extra_ideal_par1_t2
extra_ideal_par2_t2 ~~ u2*extra_ideal_par2_t2
extra_ideal_par3_t2 ~~ u3*extra_ideal_par3_t2
# Latent Variable Means
extra_ideal1 ~ 0*1
extra_ideal2 ~ 1
# Latent Variable Variances and Covariance
extra_ideal1 ~~ 1*extra_ideal1
extra_ideal2 ~~ extra_ideal2
extra_ideal1 ~~ extra_ideal2
'
fit_strict_extra_ideal <- cfa(strict_extra_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_strict_extra_ideal, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_extra_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_extra_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_extra_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_extra_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 175 19 35.1 1669. 1729. 0.910 0.831 0.139 0.0514
2 weak 175 17 36.5 1666. 1720. 0.912 0.868 0.123 0.0713
3 strong 175 15 37.5 1663. 1711. 0.915 0.894 0.110 0.0772
4 strict 175 12 39.7 1660. 1698. 0.918 0.918 0.0971 0.0985# chi-square difference test for nested models
anova(fit_configural_extra_ideal, fit_weak_extra_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_extra_ideal 8 1669.0 1729.2 35.124
fit_weak_extra_ideal 10 1666.4 1720.2 36.522 1.3981 0 2
Pr(>Chisq)
fit_configural_extra_ideal
fit_weak_extra_ideal 0.4971anova(fit_weak_extra_ideal, fit_strong_extra_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_extra_ideal 10 1666.4 1720.2 36.522
fit_strong_extra_ideal 12 1663.4 1710.9 37.517 0.99445 0 2
Pr(>Chisq)
fit_weak_extra_ideal
fit_strong_extra_ideal 0.6082anova(fit_strong_extra_ideal, fit_strict_extra_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_extra_ideal 12 1663.4 1710.9 37.517
fit_strict_extra_ideal 15 1659.6 1697.6 39.739 2.2225 0 3
Pr(>Chisq)
fit_strong_extra_ideal
fit_strict_extra_ideal 0.5275Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_agree_curr <- '
# Define the latent factors
agree_curr1 =~ NA*agree_curr_par1_t1 + lambda1*agree_curr_par1_t1 + agree_curr_par2_t1 + agree_curr_par3_t1
agree_curr2 =~ NA*agree_curr_par1_t2 + lambda1*agree_curr_par1_t2 + agree_curr_par2_t2 + agree_curr_par3_t2
# Intercepts
agree_curr_par1_t1 ~ i1*1
agree_curr_par2_t1 ~ 1
agree_curr_par3_t1 ~ 1
agree_curr_par1_t2 ~ i1*1
agree_curr_par2_t2 ~ 1
agree_curr_par3_t2 ~ 1
# Unique Variances
agree_curr_par1_t1 ~~ agree_curr_par1_t1
agree_curr_par2_t1 ~~ agree_curr_par2_t1
agree_curr_par3_t1 ~~ agree_curr_par3_t1
agree_curr_par1_t2 ~~ agree_curr_par1_t2
agree_curr_par2_t2 ~~ agree_curr_par2_t2
agree_curr_par3_t2 ~~ agree_curr_par3_t2
# Latent Variable Means
agree_curr1 ~ 0*1
agree_curr2 ~ 1
# Latent Variable Variances and Covariance
agree_curr1 ~~ 1*agree_curr1
agree_curr2 ~~ agree_curr2
agree_curr1 ~~ agree_curr2
'
fit_configural_agree_curr <- cfa(configural_agree_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_configural_agree_curr, fit.measures = TRUE)
# Weak invariance model
weak_agree_curr <- '
# Define the latent factors
agree_curr1 =~ NA*agree_curr_par1_t1 + lambda1*agree_curr_par1_t1 + lambda2*agree_curr_par2_t1 + lambda3*agree_curr_par3_t1
agree_curr2 =~ NA*agree_curr_par1_t2 + lambda1*agree_curr_par1_t2 + lambda2*agree_curr_par2_t2 + lambda3*agree_curr_par3_t2
# Intercepts
agree_curr_par1_t1 ~ i1*1
agree_curr_par2_t1 ~ 1
agree_curr_par3_t1 ~ 1
agree_curr_par1_t2 ~ i1*1
agree_curr_par2_t2 ~ 1
agree_curr_par3_t2 ~ 1
# Unique Variances
agree_curr_par1_t1 ~~ agree_curr_par1_t1
agree_curr_par2_t1 ~~ agree_curr_par2_t1
agree_curr_par3_t1 ~~ agree_curr_par3_t1
agree_curr_par1_t2 ~~ agree_curr_par1_t2
agree_curr_par2_t2 ~~ agree_curr_par2_t2
agree_curr_par3_t2 ~~ agree_curr_par3_t2
# Latent Variable Means
agree_curr1 ~ 0*1
agree_curr2 ~ 1
# Latent Variable Variances and Covariance
agree_curr1 ~~ 1*agree_curr1
agree_curr2 ~~ agree_curr2
agree_curr1 ~~ agree_curr2
'
fit_weak_agree_curr <- cfa(weak_agree_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_weak_agree_curr, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_agree_curr <- '
# Define the latent factors
agree_curr1 =~ NA*agree_curr_par1_t1 + lambda1*agree_curr_par1_t1 + lambda2*agree_curr_par2_t1 + lambda3*agree_curr_par3_t1
agree_curr2 =~ NA*agree_curr_par1_t2 + lambda1*agree_curr_par1_t2 + lambda2*agree_curr_par2_t2 + lambda3*agree_curr_par3_t2
# Intercepts
agree_curr_par1_t1 ~ i1*1
agree_curr_par2_t1 ~ i2*1
agree_curr_par3_t1 ~ i3*1
agree_curr_par1_t2 ~ i1*1
agree_curr_par2_t2 ~ i2*1
agree_curr_par3_t2 ~ i3*1
# Unique Variances
agree_curr_par1_t1 ~~ agree_curr_par1_t1
agree_curr_par2_t1 ~~ agree_curr_par2_t1
agree_curr_par3_t1 ~~ agree_curr_par3_t1
agree_curr_par1_t2 ~~ agree_curr_par1_t2
agree_curr_par2_t2 ~~ agree_curr_par2_t2
agree_curr_par3_t2 ~~ agree_curr_par3_t2
# Latent Variable Means
agree_curr1 ~ 0*1
agree_curr2 ~ 1
# Latent Variable Variances and Covariance
agree_curr1 ~~ 1*agree_curr1
agree_curr2 ~~ agree_curr2
agree_curr1 ~~ agree_curr2
'
fit_strong_agree_curr <- cfa(strong_agree_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strong_agree_curr, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_agree_curr <- '
# Define the latent factors
agree_curr1 =~ NA*agree_curr_par1_t1 + lambda1*agree_curr_par1_t1 + lambda2*agree_curr_par2_t1 + lambda3*agree_curr_par3_t1
agree_curr2 =~ NA*agree_curr_par1_t2 + lambda1*agree_curr_par1_t2 + lambda2*agree_curr_par2_t2 + lambda3*agree_curr_par3_t2
# Intercepts
agree_curr_par1_t1 ~ i1*1
agree_curr_par2_t1 ~ i2*1
agree_curr_par3_t1 ~ i3*1
agree_curr_par1_t2 ~ i1*1
agree_curr_par2_t2 ~ i2*1
agree_curr_par3_t2 ~ i3*1
# Unique Variances
agree_curr_par1_t1 ~~ u1*agree_curr_par1_t1
agree_curr_par2_t1 ~~ u2*agree_curr_par2_t1
agree_curr_par3_t1 ~~ u3*agree_curr_par3_t1
agree_curr_par1_t2 ~~ u1*agree_curr_par1_t2
agree_curr_par2_t2 ~~ u2*agree_curr_par2_t2
agree_curr_par3_t2 ~~ u3*agree_curr_par3_t2
# Latent Variable Means
agree_curr1 ~ 0*1
agree_curr2 ~ 1
# Latent Variable Variances and Covariance
agree_curr1 ~~ 1*agree_curr1
agree_curr2 ~~ agree_curr2
agree_curr1 ~~ agree_curr2
'
fit_strict_agree_curr <- cfa(strict_agree_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strict_agree_curr, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_agree_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_agree_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_agree_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_agree_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 19 241. 5004. 5085. 0.858 0.734 0.235 0.0584
2 weak 530 17 242. 5001. 5073. 0.859 0.789 0.209 0.0605
3 strong 530 15 245. 5000. 5064. 0.859 0.823 0.191 0.0624
4 strict 530 12 253. 5002. 5053. 0.856 0.856 0.173 0.0833# chi-square difference test for nested models
anova(fit_configural_agree_curr, fit_weak_agree_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_agree_curr 8 5004.2 5085.4 241.50
fit_weak_agree_curr 10 5000.8 5073.5 242.14 0.64558 0 2
Pr(>Chisq)
fit_configural_agree_curr
fit_weak_agree_curr 0.7241anova(fit_weak_agree_curr, fit_strong_agree_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_agree_curr 10 5000.8 5073.5 242.14
fit_strong_agree_curr 12 4999.6 5063.7 244.89 2.7511 0.02662 2
Pr(>Chisq)
fit_weak_agree_curr
fit_strong_agree_curr 0.2527anova(fit_strong_agree_curr, fit_strict_agree_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_agree_curr 12 4999.6 5063.7 244.89
fit_strict_agree_curr 15 5001.5 5052.8 252.81 7.9229 0.055643 3
Pr(>Chisq)
fit_strong_agree_curr
fit_strict_agree_curr 0.04763 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_agree_ideal <- '
# Define the latent factors
agree_ideal1 =~ NA*agree_ideal_par1_t1 + lambda1*agree_ideal_par1_t1 + agree_ideal_par2_t1 + agree_ideal_par3_t1
agree_ideal2 =~ NA*agree_ideal_par1_t2 + lambda1*agree_ideal_par1_t2 + agree_ideal_par2_t2 + agree_ideal_par3_t2
# Intercepts
agree_ideal_par1_t1 ~ i1*1
agree_ideal_par2_t1 ~ 1
agree_ideal_par3_t1 ~ 1
agree_ideal_par1_t2 ~ i1*1
agree_ideal_par2_t2 ~ 1
agree_ideal_par3_t2 ~ 1
# Unique Variances
agree_ideal_par1_t1 ~~ agree_ideal_par1_t1
agree_ideal_par2_t1 ~~ agree_ideal_par2_t1
agree_ideal_par3_t1 ~~ agree_ideal_par3_t1
agree_ideal_par1_t2 ~~ agree_ideal_par1_t2
agree_ideal_par2_t2 ~~ agree_ideal_par2_t2
agree_ideal_par3_t2 ~~ agree_ideal_par3_t2
# Latent Variable Means
agree_ideal1 ~ 0*1
agree_ideal2 ~ 1
# Latent Variable Variances and Covariance
agree_ideal1 ~~ 1*agree_ideal1
agree_ideal2 ~~ agree_ideal2
agree_ideal1 ~~ agree_ideal2
'
fit_configural_agree_ideal <- cfa(configural_agree_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"),
mimic = "mplus", missing="ML")
summary(fit_configural_agree_ideal, fit.measures = TRUE)
# Weak invariance model
weak_agree_ideal <- '
# Define the latent factors
agree_ideal1 =~ NA*agree_ideal_par1_t1 + lambda1*agree_ideal_par1_t1 + lambda2*agree_ideal_par2_t1 + lambda3*agree_ideal_par3_t1
agree_ideal2 =~ NA*agree_ideal_par1_t2 + lambda1*agree_ideal_par1_t2 + lambda2*agree_ideal_par2_t2 + lambda3*agree_ideal_par3_t2
# Intercepts
agree_ideal_par1_t1 ~ i1*1
agree_ideal_par2_t1 ~ 1
agree_ideal_par3_t1 ~ 1
agree_ideal_par1_t2 ~ i1*1
agree_ideal_par2_t2 ~ 1
agree_ideal_par3_t2 ~ 1
# Unique Variances
agree_ideal_par1_t1 ~~ agree_ideal_par1_t1
agree_ideal_par2_t1 ~~ agree_ideal_par2_t1
agree_ideal_par3_t1 ~~ agree_ideal_par3_t1
agree_ideal_par1_t2 ~~ agree_ideal_par1_t2
agree_ideal_par2_t2 ~~ agree_ideal_par2_t2
agree_ideal_par3_t2 ~~ agree_ideal_par3_t2
# Latent Variable Means
agree_ideal1 ~ 0*1
agree_ideal2 ~ 1
# Latent Variable Variances and Covariance
agree_ideal1 ~~ 1*agree_ideal1
agree_ideal2 ~~ agree_ideal2
agree_ideal1 ~~ agree_ideal2
'
fit_weak_agree_ideal <- cfa(weak_agree_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_weak_agree_ideal, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_agree_ideal <- '
# Define the latent factors
agree_ideal1 =~ NA*agree_ideal_par1_t1 + lambda1*agree_ideal_par1_t1 + lambda2*agree_ideal_par2_t1 + lambda3*agree_ideal_par3_t1
agree_ideal2 =~ NA*agree_ideal_par1_t2 + lambda1*agree_ideal_par1_t2 + lambda2*agree_ideal_par2_t2 + lambda3*agree_ideal_par3_t2
# Intercepts
agree_ideal_par1_t1 ~ i1*1
agree_ideal_par2_t1 ~ i2*1
agree_ideal_par3_t1 ~ i3*1
agree_ideal_par1_t2 ~ i1*1
agree_ideal_par2_t2 ~ i2*1
agree_ideal_par3_t2 ~ i3*1
# Unique Variances
agree_ideal_par1_t1 ~~ agree_ideal_par1_t1
agree_ideal_par2_t1 ~~ agree_ideal_par2_t1
agree_ideal_par3_t1 ~~ agree_ideal_par3_t1
agree_ideal_par1_t2 ~~ agree_ideal_par1_t2
agree_ideal_par2_t2 ~~ agree_ideal_par2_t2
agree_ideal_par3_t2 ~~ agree_ideal_par3_t2
# Latent Variable Means
agree_ideal1 ~ 0*1
agree_ideal2 ~ 1
# Latent Variable Variances and Covariance
agree_ideal1 ~~ 1*agree_ideal1
agree_ideal2 ~~ agree_ideal2
agree_ideal1 ~~ agree_ideal2
'
fit_strong_agree_ideal <- cfa(strong_agree_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_strong_agree_ideal, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_agree_ideal <- '
# Define the latent factors
agree_ideal1 =~ NA*agree_ideal_par1_t1 + lambda1*agree_ideal_par1_t1 + lambda2*agree_ideal_par2_t1 + lambda3*agree_ideal_par3_t1
agree_ideal2 =~ NA*agree_ideal_par1_t2 + lambda1*agree_ideal_par1_t2 + lambda2*agree_ideal_par2_t2 + lambda3*agree_ideal_par3_t2
# Intercepts
agree_ideal_par1_t1 ~ i1*1
agree_ideal_par2_t1 ~ i2*1
agree_ideal_par3_t1 ~ i3*1
agree_ideal_par1_t2 ~ i1*1
agree_ideal_par2_t2 ~ i2*1
agree_ideal_par3_t2 ~ i3*1
# Unique Variances
agree_ideal_par1_t1 ~~ u1*agree_ideal_par1_t1
agree_ideal_par2_t1 ~~ u2*agree_ideal_par2_t1
agree_ideal_par3_t1 ~~ u3*agree_ideal_par3_t1
agree_ideal_par1_t2 ~~ u1*agree_ideal_par1_t2
agree_ideal_par2_t2 ~~ u2*agree_ideal_par2_t2
agree_ideal_par3_t2 ~~ u3*agree_ideal_par3_t2
# Latent Variable Means
agree_ideal1 ~ 0*1
agree_ideal2 ~ 1
# Latent Variable Variances and Covariance
agree_ideal1 ~~ 1*agree_ideal1
agree_ideal2 ~~ agree_ideal2
agree_ideal1 ~~ agree_ideal2
'
fit_strict_agree_ideal <- cfa(strict_agree_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_strict_agree_ideal, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_agree_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_agree_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_agree_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_agree_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 175 19 36.4 1606. 1666. 0.923 0.856 0.142 0.0471
2 weak 175 17 38.5 1604. 1658. 0.923 0.884 0.128 0.0786
3 strong 175 15 41.0 1603. 1650. 0.921 0.902 0.118 0.0803
4 strict 175 12 44.2 1600. 1638. 0.921 0.921 0.105 0.100 # chi-square difference test for nested models
anova(fit_configural_agree_ideal, fit_weak_agree_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_agree_ideal 8 1606.2 1666.3 36.375
fit_weak_agree_ideal 10 1604.3 1658.1 38.507 2.1316 0.019388 2
Pr(>Chisq)
fit_configural_agree_ideal
fit_weak_agree_ideal 0.3445anova(fit_weak_agree_ideal, fit_strong_agree_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_agree_ideal 10 1604.3 1658.1 38.507
fit_strong_agree_ideal 12 1602.8 1650.3 41.024 2.5174 0.03845 2
Pr(>Chisq)
fit_weak_agree_ideal
fit_strong_agree_ideal 0.284anova(fit_strong_agree_ideal, fit_strict_agree_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_agree_ideal 12 1602.8 1650.3 41.024
fit_strict_agree_ideal 15 1600.0 1638.0 44.200 3.1753 0.018273 3
Pr(>Chisq)
fit_strong_agree_ideal
fit_strict_agree_ideal 0.3654Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_consc_curr <- '
# Define the latent factors
consc_curr1 =~ NA*consc_curr_par1_t1 + lambda1*consc_curr_par1_t1 + consc_curr_par2_t1 + consc_curr_par3_t1
consc_curr2 =~ NA*consc_curr_par1_t2 + lambda1*consc_curr_par1_t2 + consc_curr_par2_t2 + consc_curr_par3_t2
# Intercepts
consc_curr_par1_t1 ~ i1*1
consc_curr_par2_t1 ~ 1
consc_curr_par3_t1 ~ 1
consc_curr_par1_t2 ~ i1*1
consc_curr_par2_t2 ~ 1
consc_curr_par3_t2 ~ 1
# Unique Variances
consc_curr_par1_t1 ~~ consc_curr_par1_t1
consc_curr_par2_t1 ~~ consc_curr_par2_t1
consc_curr_par3_t1 ~~ consc_curr_par3_t1
consc_curr_par1_t2 ~~ consc_curr_par1_t2
consc_curr_par2_t2 ~~ consc_curr_par2_t2
consc_curr_par3_t2 ~~ consc_curr_par3_t2
# Latent Variable Means
consc_curr1 ~ 0*1
consc_curr2 ~ 1
# Latent Variable Variances and Covariance
consc_curr1 ~~ 1*consc_curr1
consc_curr2 ~~ consc_curr2
consc_curr1 ~~ consc_curr2
'
fit_configural_consc_curr <- cfa(configural_consc_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_configural_consc_curr, fit.measures = TRUE)
# Weak invariance model
weak_consc_curr <- '
# Define the latent factors
consc_curr1 =~ NA*consc_curr_par1_t1 + lambda1*consc_curr_par1_t1 + lambda2*consc_curr_par2_t1 + lambda3*consc_curr_par3_t1
consc_curr2 =~ NA*consc_curr_par1_t2 + lambda1*consc_curr_par1_t2 + lambda2*consc_curr_par2_t2 + lambda3*consc_curr_par3_t2
# Intercepts
consc_curr_par1_t1 ~ i1*1
consc_curr_par2_t1 ~ 1
consc_curr_par3_t1 ~ 1
consc_curr_par1_t2 ~ i1*1
consc_curr_par2_t2 ~ 1
consc_curr_par3_t2 ~ 1
# Unique Variances
consc_curr_par1_t1 ~~ consc_curr_par1_t1
consc_curr_par2_t1 ~~ consc_curr_par2_t1
consc_curr_par3_t1 ~~ consc_curr_par3_t1
consc_curr_par1_t2 ~~ consc_curr_par1_t2
consc_curr_par2_t2 ~~ consc_curr_par2_t2
consc_curr_par3_t2 ~~ consc_curr_par3_t2
# Latent Variable Means
consc_curr1 ~ 0*1
consc_curr2 ~ 1
# Latent Variable Variances and Covariance
consc_curr1 ~~ 1*consc_curr1
consc_curr2 ~~ consc_curr2
consc_curr1 ~~ consc_curr2
'
fit_weak_consc_curr <- cfa(weak_consc_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_weak_consc_curr, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_consc_curr <- '
# Define the latent factors
consc_curr1 =~ NA*consc_curr_par1_t1 + lambda1*consc_curr_par1_t1 + lambda2*consc_curr_par2_t1 + lambda3*consc_curr_par3_t1
consc_curr2 =~ NA*consc_curr_par1_t2 + lambda1*consc_curr_par1_t2 + lambda2*consc_curr_par2_t2 + lambda3*consc_curr_par3_t2
# Intercepts
consc_curr_par1_t1 ~ i1*1
consc_curr_par2_t1 ~ i2*1
consc_curr_par3_t1 ~ i3*1
consc_curr_par1_t2 ~ i1*1
consc_curr_par2_t2 ~ i2*1
consc_curr_par3_t2 ~ i3*1
# Unique Variances
consc_curr_par1_t1 ~~ consc_curr_par1_t1
consc_curr_par2_t1 ~~ consc_curr_par2_t1
consc_curr_par3_t1 ~~ consc_curr_par3_t1
consc_curr_par1_t2 ~~ consc_curr_par1_t2
consc_curr_par2_t2 ~~ consc_curr_par2_t2
consc_curr_par3_t2 ~~ consc_curr_par3_t2
# Latent Variable Means
consc_curr1 ~ 0*1
consc_curr2 ~ 1
# Latent Variable Variances and Covariance
consc_curr1 ~~ 1*consc_curr1
consc_curr2 ~~ consc_curr2
consc_curr1 ~~ consc_curr2
'
fit_strong_consc_curr <- cfa(strong_consc_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strong_consc_curr, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_consc_curr <- '
# Define the latent factors
consc_curr1 =~ NA*consc_curr_par1_t1 + lambda1*consc_curr_par1_t1 + lambda2*consc_curr_par2_t1 + lambda3*consc_curr_par3_t1
consc_curr2 =~ NA*consc_curr_par1_t2 + lambda1*consc_curr_par1_t2 + lambda2*consc_curr_par2_t2 + lambda3*consc_curr_par3_t2
# Intercepts
consc_curr_par1_t1 ~ i1*1
consc_curr_par2_t1 ~ i2*1
consc_curr_par3_t1 ~ i3*1
consc_curr_par1_t2 ~ i1*1
consc_curr_par2_t2 ~ i2*1
consc_curr_par3_t2 ~ i3*1
# Unique Variances
consc_curr_par1_t1 ~~ u1*consc_curr_par1_t1
consc_curr_par2_t1 ~~ u2*consc_curr_par2_t1
consc_curr_par3_t1 ~~ u3*consc_curr_par3_t1
consc_curr_par1_t2 ~~ u1*consc_curr_par1_t2
consc_curr_par2_t2 ~~ u2*consc_curr_par2_t2
consc_curr_par3_t2 ~~ u3*consc_curr_par3_t2
# Latent Variable Means
consc_curr1 ~ 0*1
consc_curr2 ~ 1
# Latent Variable Variances and Covariance
consc_curr1 ~~ 1*consc_curr1
consc_curr2 ~~ consc_curr2
consc_curr1 ~~ consc_curr2
'
fit_strict_consc_curr <- cfa(strict_consc_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strict_consc_curr, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_consc_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_consc_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_consc_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_consc_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 19 207. 5035. 5116. 0.921 0.851 0.217 0.0358
2 weak 530 17 207. 5031. 5103. 0.921 0.882 0.193 0.0359
3 strong 530 15 208. 5027. 5092. 0.922 0.902 0.175 0.0358
4 strict 530 12 211. 5024. 5076. 0.922 0.922 0.157 0.0401# chi-square difference test for nested models
anova(fit_configural_consc_curr, fit_weak_consc_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_consc_curr 8 5034.5 5115.7 206.85
fit_weak_consc_curr 10 5030.7 5103.4 207.05 0.20391 0 2
Pr(>Chisq)
fit_configural_consc_curr
fit_weak_consc_curr 0.9031anova(fit_weak_consc_curr, fit_strong_consc_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_consc_curr 10 5030.7 5103.4 207.05
fit_strong_consc_curr 12 5027.5 5091.6 207.80 0.74741 0 2
Pr(>Chisq)
fit_weak_consc_curr
fit_strong_consc_curr 0.6882anova(fit_strong_consc_curr, fit_strict_consc_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_consc_curr 12 5027.5 5091.6 207.80
fit_strict_consc_curr 15 5024.2 5075.5 210.55 2.7538 0 3
Pr(>Chisq)
fit_strong_consc_curr
fit_strict_consc_curr 0.4312Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_consc_ideal <- '
# Define the latent factors
consc_ideal1 =~ NA*consc_ideal_par1_t1 + lambda1*consc_ideal_par1_t1 + consc_ideal_par2_t1 + consc_ideal_par3_t1
consc_ideal2 =~ NA*consc_ideal_par1_t2 + lambda1*consc_ideal_par1_t2 + consc_ideal_par2_t2 + consc_ideal_par3_t2
# Intercepts
consc_ideal_par1_t1 ~ i1*1
consc_ideal_par2_t1 ~ 1
consc_ideal_par3_t1 ~ 1
consc_ideal_par1_t2 ~ i1*1
consc_ideal_par2_t2 ~ 1
consc_ideal_par3_t2 ~ 1
# Unique Variances
consc_ideal_par1_t1 ~~ consc_ideal_par1_t1
consc_ideal_par2_t1 ~~ consc_ideal_par2_t1
consc_ideal_par3_t1 ~~ consc_ideal_par3_t1
consc_ideal_par1_t2 ~~ consc_ideal_par1_t2
consc_ideal_par2_t2 ~~ consc_ideal_par2_t2
consc_ideal_par3_t2 ~~ consc_ideal_par3_t2
# Latent Variable Means
consc_ideal1 ~ 0*1
consc_ideal2 ~ 1
# Latent Variable Variances and Covariance
consc_ideal1 ~~ 1*consc_ideal1
consc_ideal2 ~~ consc_ideal2
consc_ideal1 ~~ consc_ideal2
'
fit_configural_consc_ideal <- cfa(configural_consc_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_configural_consc_ideal, fit.measures = TRUE)
# Weak invariance model
weak_consc_ideal <- '
# Define the latent factors
consc_ideal1 =~ NA*consc_ideal_par1_t1 + lambda1*consc_ideal_par1_t1 + lambda2*consc_ideal_par2_t1 + lambda3*consc_ideal_par3_t1
consc_ideal2 =~ NA*consc_ideal_par1_t2 + lambda1*consc_ideal_par1_t2 + lambda2*consc_ideal_par2_t2 + lambda3*consc_ideal_par3_t2
# Intercepts
consc_ideal_par1_t1 ~ i1*1
consc_ideal_par2_t1 ~ 1
consc_ideal_par3_t1 ~ 1
consc_ideal_par1_t2 ~ i1*1
consc_ideal_par2_t2 ~ 1
consc_ideal_par3_t2 ~ 1
# Unique Variances
consc_ideal_par1_t1 ~~ consc_ideal_par1_t1
consc_ideal_par2_t1 ~~ consc_ideal_par2_t1
consc_ideal_par3_t1 ~~ consc_ideal_par3_t1
consc_ideal_par1_t2 ~~ consc_ideal_par1_t2
consc_ideal_par2_t2 ~~ consc_ideal_par2_t2
consc_ideal_par3_t2 ~~ consc_ideal_par3_t2
# Latent Variable Means
consc_ideal1 ~ 0*1
consc_ideal2 ~ 1
# Latent Variable Variances and Covariance
consc_ideal1 ~~ 1*consc_ideal1
consc_ideal2 ~~ consc_ideal2
consc_ideal1 ~~ consc_ideal2
'
fit_weak_consc_ideal <- cfa(weak_consc_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_weak_consc_ideal, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_consc_ideal <- '
# Define the latent factors
consc_ideal1 =~ NA*consc_ideal_par1_t1 + lambda1*consc_ideal_par1_t1 + lambda2*consc_ideal_par2_t1 + lambda3*consc_ideal_par3_t1
consc_ideal2 =~ NA*consc_ideal_par1_t2 + lambda1*consc_ideal_par1_t2 + lambda2*consc_ideal_par2_t2 + lambda3*consc_ideal_par3_t2
# Intercepts
consc_ideal_par1_t1 ~ i1*1
consc_ideal_par2_t1 ~ i2*1
consc_ideal_par3_t1 ~ i3*1
consc_ideal_par1_t2 ~ i1*1
consc_ideal_par2_t2 ~ i2*1
consc_ideal_par3_t2 ~ i3*1
# Unique Variances
consc_ideal_par1_t1 ~~ consc_ideal_par1_t1
consc_ideal_par2_t1 ~~ consc_ideal_par2_t1
consc_ideal_par3_t1 ~~ consc_ideal_par3_t1
consc_ideal_par1_t2 ~~ consc_ideal_par1_t2
consc_ideal_par2_t2 ~~ consc_ideal_par2_t2
consc_ideal_par3_t2 ~~ consc_ideal_par3_t2
# Latent Variable Means
consc_ideal1 ~ 0*1
consc_ideal2 ~ 1
# Latent Variable Variances and Covariance
consc_ideal1 ~~ 1*consc_ideal1
consc_ideal2 ~~ consc_ideal2
consc_ideal1 ~~ consc_ideal2
'
fit_strong_consc_ideal <- cfa(strong_consc_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_strong_consc_ideal, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_consc_ideal <- '
# Define the latent factors
consc_ideal1 =~ NA*consc_ideal_par1_t1 + lambda1*consc_ideal_par1_t1 + lambda2*consc_ideal_par2_t1 + lambda3*consc_ideal_par3_t1
consc_ideal2 =~ NA*consc_ideal_par1_t2 + lambda1*consc_ideal_par1_t2 + lambda2*consc_ideal_par2_t2 + lambda3*consc_ideal_par3_t2
# Intercepts
consc_ideal_par1_t1 ~ i1*1
consc_ideal_par2_t1 ~ i2*1
consc_ideal_par3_t1 ~ i3*1
consc_ideal_par1_t2 ~ i1*1
consc_ideal_par2_t2 ~ i2*1
consc_ideal_par3_t2 ~ i3*1
# Unique Variances
consc_ideal_par1_t1 ~~ u1*consc_ideal_par1_t1
consc_ideal_par2_t1 ~~ u2*consc_ideal_par2_t1
consc_ideal_par3_t1 ~~ u3*consc_ideal_par3_t1
consc_ideal_par1_t2 ~~ u1*consc_ideal_par1_t2
consc_ideal_par2_t2 ~~ u2*consc_ideal_par2_t2
consc_ideal_par3_t2 ~~ u3*consc_ideal_par3_t2
# Latent Variable Means
consc_ideal1 ~ 0*1
consc_ideal2 ~ 1
# Latent Variable Variances and Covariance
consc_ideal1 ~~ 1*consc_ideal1
consc_ideal2 ~~ consc_ideal2
consc_ideal1 ~~ consc_ideal2
'
fit_strict_consc_ideal <- cfa(strict_consc_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_strict_consc_ideal, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_consc_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_consc_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_consc_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_consc_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 175 19 27.7 1361. 1422. 0.930 0.870 0.119 0.0437
2 weak 175 17 28.7 1358. 1412. 0.934 0.901 0.103 0.0647
3 strong 175 15 29.0 1355. 1402. 0.940 0.925 0.0899 0.0685
4 strict 175 12 31.1 1351. 1389. 0.943 0.943 0.0782 0.0858# chi-square difference test for nested models
anova(fit_configural_consc_ideal, fit_weak_consc_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_consc_ideal 8 1361.4 1421.5 27.747
fit_weak_consc_ideal 10 1358.4 1412.2 28.737 0.99026 0 2
Pr(>Chisq)
fit_configural_consc_ideal
fit_weak_consc_ideal 0.6095anova(fit_weak_consc_ideal, fit_strong_consc_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_consc_ideal 10 1358.4 1412.2 28.737
fit_strong_consc_ideal 12 1354.6 1402.1 28.981 0.24429 0 2
Pr(>Chisq)
fit_weak_consc_ideal
fit_strong_consc_ideal 0.885anova(fit_strong_consc_ideal, fit_strict_consc_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_consc_ideal 12 1354.6 1402.1 28.981
fit_strict_consc_ideal 15 1350.7 1388.7 31.062 2.0812 0 3
Pr(>Chisq)
fit_strong_consc_ideal
fit_strict_consc_ideal 0.5557Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_neuro_curr <- '
# Define the latent factors
neuro_curr1 =~ NA*neuro_curr_par1_t1 + lambda1*neuro_curr_par1_t1 + neuro_curr_par2_t1 + neuro_curr_par3_t1
neuro_curr2 =~ NA*neuro_curr_par1_t2 + lambda1*neuro_curr_par1_t2 + neuro_curr_par2_t2 + neuro_curr_par3_t2
# Intercepts
neuro_curr_par1_t1 ~ i1*1
neuro_curr_par2_t1 ~ 1
neuro_curr_par3_t1 ~ 1
neuro_curr_par1_t2 ~ i1*1
neuro_curr_par2_t2 ~ 1
neuro_curr_par3_t2 ~ 1
# Unique Variances
neuro_curr_par1_t1 ~~ neuro_curr_par1_t1
neuro_curr_par2_t1 ~~ neuro_curr_par2_t1
neuro_curr_par3_t1 ~~ neuro_curr_par3_t1
neuro_curr_par1_t2 ~~ neuro_curr_par1_t2
neuro_curr_par2_t2 ~~ neuro_curr_par2_t2
neuro_curr_par3_t2 ~~ neuro_curr_par3_t2
# Latent Variable Means
neuro_curr1 ~ 0*1
neuro_curr2 ~ 1
# Latent Variable Variances and Covariance
neuro_curr1 ~~ 1*neuro_curr1
neuro_curr2 ~~ neuro_curr2
neuro_curr1 ~~ neuro_curr2
'
fit_configural_neuro_curr <- cfa(configural_neuro_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_configural_neuro_curr, fit.measures = TRUE)
# Weak invariance model
weak_neuro_curr <- '
# Define the latent factors
neuro_curr1 =~ NA*neuro_curr_par1_t1 + lambda1*neuro_curr_par1_t1 + lambda2*neuro_curr_par2_t1 + lambda3*neuro_curr_par3_t1
neuro_curr2 =~ NA*neuro_curr_par1_t2 + lambda1*neuro_curr_par1_t2 + lambda2*neuro_curr_par2_t2 + lambda3*neuro_curr_par3_t2
# Intercepts
neuro_curr_par1_t1 ~ i1*1
neuro_curr_par2_t1 ~ 1
neuro_curr_par3_t1 ~ 1
neuro_curr_par1_t2 ~ i1*1
neuro_curr_par2_t2 ~ 1
neuro_curr_par3_t2 ~ 1
# Unique Variances
neuro_curr_par1_t1 ~~ neuro_curr_par1_t1
neuro_curr_par2_t1 ~~ neuro_curr_par2_t1
neuro_curr_par3_t1 ~~ neuro_curr_par3_t1
neuro_curr_par1_t2 ~~ neuro_curr_par1_t2
neuro_curr_par2_t2 ~~ neuro_curr_par2_t2
neuro_curr_par3_t2 ~~ neuro_curr_par3_t2
# Latent Variable Means
neuro_curr1 ~ 0*1
neuro_curr2 ~ 1
# Latent Variable Variances and Covariance
neuro_curr1 ~~ 1*neuro_curr1
neuro_curr2 ~~ neuro_curr2
neuro_curr1 ~~ neuro_curr2
'
fit_weak_neuro_curr <- cfa(weak_neuro_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_weak_neuro_curr, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_neuro_curr <- '
# Define the latent factors
neuro_curr1 =~ NA*neuro_curr_par1_t1 + lambda1*neuro_curr_par1_t1 + lambda2*neuro_curr_par2_t1 + lambda3*neuro_curr_par3_t1
neuro_curr2 =~ NA*neuro_curr_par1_t2 + lambda1*neuro_curr_par1_t2 + lambda2*neuro_curr_par2_t2 + lambda3*neuro_curr_par3_t2
# Intercepts
neuro_curr_par1_t1 ~ i1*1
neuro_curr_par2_t1 ~ i2*1
neuro_curr_par3_t1 ~ i3*1
neuro_curr_par1_t2 ~ i1*1
neuro_curr_par2_t2 ~ i2*1
neuro_curr_par3_t2 ~ i3*1
# Unique Variances
neuro_curr_par1_t1 ~~ neuro_curr_par1_t1
neuro_curr_par2_t1 ~~ neuro_curr_par2_t1
neuro_curr_par3_t1 ~~ neuro_curr_par3_t1
neuro_curr_par1_t2 ~~ neuro_curr_par1_t2
neuro_curr_par2_t2 ~~ neuro_curr_par2_t2
neuro_curr_par3_t2 ~~ neuro_curr_par3_t2
# Latent Variable Means
neuro_curr1 ~ 0*1
neuro_curr2 ~ 1
# Latent Variable Variances and Covariance
neuro_curr1 ~~ 1*neuro_curr1
neuro_curr2 ~~ neuro_curr2
neuro_curr1 ~~ neuro_curr2
'
fit_strong_neuro_curr <- cfa(strong_neuro_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strong_neuro_curr, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_neuro_curr <- '
# Define the latent factors
neuro_curr1 =~ NA*neuro_curr_par1_t1 + lambda1*neuro_curr_par1_t1 + lambda2*neuro_curr_par2_t1 + lambda3*neuro_curr_par3_t1
neuro_curr2 =~ NA*neuro_curr_par1_t2 + lambda1*neuro_curr_par1_t2 + lambda2*neuro_curr_par2_t2 + lambda3*neuro_curr_par3_t2
# Intercepts
neuro_curr_par1_t1 ~ i1*1
neuro_curr_par2_t1 ~ i2*1
neuro_curr_par3_t1 ~ i3*1
neuro_curr_par1_t2 ~ i1*1
neuro_curr_par2_t2 ~ i2*1
neuro_curr_par3_t2 ~ i3*1
# Unique Variances
neuro_curr_par1_t1 ~~ u1*neuro_curr_par1_t1
neuro_curr_par2_t1 ~~ u2*neuro_curr_par2_t1
neuro_curr_par3_t1 ~~ u3*neuro_curr_par3_t1
neuro_curr_par1_t2 ~~ u1*neuro_curr_par1_t2
neuro_curr_par2_t2 ~~ u2*neuro_curr_par2_t2
neuro_curr_par3_t2 ~~ u3*neuro_curr_par3_t2
# Latent Variable Means
neuro_curr1 ~ 0*1
neuro_curr2 ~ 1
# Latent Variable Variances and Covariance
neuro_curr1 ~~ 1*neuro_curr1
neuro_curr2 ~~ neuro_curr2
neuro_curr1 ~~ neuro_curr2
'
fit_strict_neuro_curr <- cfa(strict_neuro_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strict_neuro_curr, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_neuro_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_neuro_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_neuro_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_neuro_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 19 141. 5055. 5136. 0.959 0.923 0.177 0.0213
2 weak 530 17 142. 5052. 5124. 0.959 0.939 0.158 0.0238
3 strong 530 15 149. 5055. 5119. 0.958 0.947 0.147 0.0231
4 strict 530 12 151. 5051. 5102. 0.958 0.958 0.131 0.0242# chi-square difference test for nested models
anova(fit_configural_neuro_curr, fit_weak_neuro_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_neuro_curr 8 5054.9 5136.1 141.36
fit_weak_neuro_curr 10 5051.5 5124.2 141.97 0.60831 0 2
Pr(>Chisq)
fit_configural_neuro_curr
fit_weak_neuro_curr 0.7377anova(fit_weak_neuro_curr, fit_strong_neuro_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_neuro_curr 10 5051.5 5124.2 141.97
fit_strong_neuro_curr 12 5054.9 5119.0 149.40 7.4313 0.071581 2
Pr(>Chisq)
fit_weak_neuro_curr
fit_strong_neuro_curr 0.02434 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1anova(fit_strong_neuro_curr, fit_strict_neuro_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_neuro_curr 12 5054.9 5119.0 149.40
fit_strict_neuro_curr 15 5051.0 5102.3 151.44 2.0399 0 3
Pr(>Chisq)
fit_strong_neuro_curr
fit_strict_neuro_curr 0.5642Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_neuro_ideal <- '
# Define the latent factors
neuro_ideal1 =~ NA*neuro_ideal_par1_t1 + lambda1*neuro_ideal_par1_t1 + neuro_ideal_par2_t1 + neuro_ideal_par3_t1
neuro_ideal2 =~ NA*neuro_ideal_par1_t2 + lambda1*neuro_ideal_par1_t2 + neuro_ideal_par2_t2 + neuro_ideal_par3_t2
# Intercepts
neuro_ideal_par1_t1 ~ i1*1
neuro_ideal_par2_t1 ~ 1
neuro_ideal_par3_t1 ~ 1
neuro_ideal_par1_t2 ~ i1*1
neuro_ideal_par2_t2 ~ 1
neuro_ideal_par3_t2 ~ 1
# Unique Variances
neuro_ideal_par1_t1 ~~ neuro_ideal_par1_t1
neuro_ideal_par2_t1 ~~ neuro_ideal_par2_t1
neuro_ideal_par3_t1 ~~ neuro_ideal_par3_t1
neuro_ideal_par1_t2 ~~ neuro_ideal_par1_t2
neuro_ideal_par2_t2 ~~ neuro_ideal_par2_t2
neuro_ideal_par3_t2 ~~ neuro_ideal_par3_t2
# Latent Variable Means
neuro_ideal1 ~ 0*1
neuro_ideal2 ~ 1
# Latent Variable Variances and Covariance
neuro_ideal1 ~~ 1*neuro_ideal1
neuro_ideal2 ~~ neuro_ideal2
neuro_ideal1 ~~ neuro_ideal2
'
fit_configural_neuro_ideal <- cfa(configural_neuro_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_configural_neuro_ideal, fit.measures = TRUE)
# Weak invariance model
weak_neuro_ideal <- '
# Define the latent factors
neuro_ideal1 =~ NA*neuro_ideal_par1_t1 + lambda1*neuro_ideal_par1_t1 + lambda2*neuro_ideal_par2_t1 + lambda3*neuro_ideal_par3_t1
neuro_ideal2 =~ NA*neuro_ideal_par1_t2 + lambda1*neuro_ideal_par1_t2 + lambda2*neuro_ideal_par2_t2 + lambda3*neuro_ideal_par3_t2
# Intercepts
neuro_ideal_par1_t1 ~ i1*1
neuro_ideal_par2_t1 ~ 1
neuro_ideal_par3_t1 ~ 1
neuro_ideal_par1_t2 ~ i1*1
neuro_ideal_par2_t2 ~ 1
neuro_ideal_par3_t2 ~ 1
# Unique Variances
neuro_ideal_par1_t1 ~~ neuro_ideal_par1_t1
neuro_ideal_par2_t1 ~~ neuro_ideal_par2_t1
neuro_ideal_par3_t1 ~~ neuro_ideal_par3_t1
neuro_ideal_par1_t2 ~~ neuro_ideal_par1_t2
neuro_ideal_par2_t2 ~~ neuro_ideal_par2_t2
neuro_ideal_par3_t2 ~~ neuro_ideal_par3_t2
# Latent Variable Means
neuro_ideal1 ~ 0*1
neuro_ideal2 ~ 1
# Latent Variable Variances and Covariance
neuro_ideal1 ~~ 1*neuro_ideal1
neuro_ideal2 ~~ neuro_ideal2
neuro_ideal1 ~~ neuro_ideal2
'
fit_weak_neuro_ideal <- cfa(weak_neuro_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_weak_neuro_ideal, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_neuro_ideal <- '
# Define the latent factors
neuro_ideal1 =~ NA*neuro_ideal_par1_t1 + lambda1*neuro_ideal_par1_t1 + lambda2*neuro_ideal_par2_t1 + lambda3*neuro_ideal_par3_t1
neuro_ideal2 =~ NA*neuro_ideal_par1_t2 + lambda1*neuro_ideal_par1_t2 + lambda2*neuro_ideal_par2_t2 + lambda3*neuro_ideal_par3_t2
# Intercepts
neuro_ideal_par1_t1 ~ i1*1
neuro_ideal_par2_t1 ~ i2*1
neuro_ideal_par3_t1 ~ i3*1
neuro_ideal_par1_t2 ~ i1*1
neuro_ideal_par2_t2 ~ i2*1
neuro_ideal_par3_t2 ~ i3*1
# Unique Variances
neuro_ideal_par1_t1 ~~ neuro_ideal_par1_t1
neuro_ideal_par2_t1 ~~ neuro_ideal_par2_t1
neuro_ideal_par3_t1 ~~ neuro_ideal_par3_t1
neuro_ideal_par1_t2 ~~ neuro_ideal_par1_t2
neuro_ideal_par2_t2 ~~ neuro_ideal_par2_t2
neuro_ideal_par3_t2 ~~ neuro_ideal_par3_t2
# Latent Variable Means
neuro_ideal1 ~ 0*1
neuro_ideal2 ~ 1
# Latent Variable Variances and Covariance
neuro_ideal1 ~~ 1*neuro_ideal1
neuro_ideal2 ~~ neuro_ideal2
neuro_ideal1 ~~ neuro_ideal2
'
fit_strong_neuro_ideal <- cfa(strong_neuro_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_strong_neuro_ideal, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_neuro_ideal <- '
# Define the latent factors
neuro_ideal1 =~ NA*neuro_ideal_par1_t1 + lambda1*neuro_ideal_par1_t1 + lambda2*neuro_ideal_par2_t1 + lambda3*neuro_ideal_par3_t1
neuro_ideal2 =~ NA*neuro_ideal_par1_t2 + lambda1*neuro_ideal_par1_t2 + lambda2*neuro_ideal_par2_t2 + lambda3*neuro_ideal_par3_t2
# Intercepts
neuro_ideal_par1_t1 ~ i1*1
neuro_ideal_par2_t1 ~ i2*1
neuro_ideal_par3_t1 ~ i3*1
neuro_ideal_par1_t2 ~ i1*1
neuro_ideal_par2_t2 ~ i2*1
neuro_ideal_par3_t2 ~ i3*1
# Unique Variances
neuro_ideal_par1_t1 ~~ u1*neuro_ideal_par1_t1
neuro_ideal_par2_t1 ~~ u2*neuro_ideal_par2_t1
neuro_ideal_par3_t1 ~~ u3*neuro_ideal_par3_t1
neuro_ideal_par1_t2 ~~ u1*neuro_ideal_par1_t2
neuro_ideal_par2_t2 ~~ u2*neuro_ideal_par2_t2
neuro_ideal_par3_t2 ~~ u3*neuro_ideal_par3_t2
# Latent Variable Means
neuro_ideal1 ~ 0*1
neuro_ideal2 ~ 1
# Latent Variable Variances and Covariance
neuro_ideal1 ~~ 1*neuro_ideal1
neuro_ideal2 ~~ neuro_ideal2
neuro_ideal1 ~~ neuro_ideal2
'
fit_strict_neuro_ideal <- cfa(strict_neuro_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_strict_neuro_ideal, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_neuro_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_neuro_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_neuro_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_neuro_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 175 19 10.6 1373. 1433. 0.990 0.982 0.0427 0.0314
2 weak 175 17 10.6 1369. 1423. 0.998 0.997 0.0180 0.0312
3 strong 175 15 19.5 1374. 1421. 0.972 0.965 0.0598 0.0410
4 strict 175 12 26.8 1375. 1413. 0.955 0.955 0.0671 0.0770# chi-square difference test for nested models
anova(fit_configural_neuro_ideal, fit_weak_neuro_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_neuro_ideal 8 1372.8 1432.9 10.552
fit_weak_neuro_ideal 10 1368.8 1422.6 10.570 0.018348 0 2
Pr(>Chisq)
fit_configural_neuro_ideal
fit_weak_neuro_ideal 0.9909anova(fit_weak_neuro_ideal, fit_strong_neuro_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_neuro_ideal 10 1368.8 1422.6 10.570
fit_strong_neuro_ideal 12 1373.8 1421.2 19.515 8.9452 0.14087 2
Pr(>Chisq)
fit_weak_neuro_ideal
fit_strong_neuro_ideal 0.01142 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1anova(fit_strong_neuro_ideal, fit_strict_neuro_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_neuro_ideal 12 1373.8 1421.2 19.515
fit_strict_neuro_ideal 15 1375.1 1413.0 26.815 7.2996 0.090497 3
Pr(>Chisq)
fit_strong_neuro_ideal
fit_strict_neuro_ideal 0.06294 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_openn_curr <- '
# Define the latent factors
openn_curr1 =~ NA*openn_curr_par1_t1 + lambda1*openn_curr_par1_t1 + openn_curr_par2_t1 + openn_curr_par3_t1
openn_curr2 =~ NA*openn_curr_par1_t2 + lambda1*openn_curr_par1_t2 + openn_curr_par2_t2 + openn_curr_par3_t2
# Intercepts
openn_curr_par1_t1 ~ i1*1
openn_curr_par2_t1 ~ 1
openn_curr_par3_t1 ~ 1
openn_curr_par1_t2 ~ i1*1
openn_curr_par2_t2 ~ 1
openn_curr_par3_t2 ~ 1
# Unique Variances
openn_curr_par1_t1 ~~ openn_curr_par1_t1
openn_curr_par2_t1 ~~ openn_curr_par2_t1
openn_curr_par3_t1 ~~ openn_curr_par3_t1
openn_curr_par1_t2 ~~ openn_curr_par1_t2
openn_curr_par2_t2 ~~ openn_curr_par2_t2
openn_curr_par3_t2 ~~ openn_curr_par3_t2
# Latent Variable Means
openn_curr1 ~ 0*1
openn_curr2 ~ 1
# Latent Variable Variances and Covariance
openn_curr1 ~~ 1*openn_curr1
openn_curr2 ~~ openn_curr2
openn_curr1 ~~ openn_curr2
'
fit_configural_openn_curr <- cfa(configural_openn_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_configural_openn_curr, fit.measures = TRUE)
# Weak invariance model
weak_openn_curr <- '
# Define the latent factors
openn_curr1 =~ NA*openn_curr_par1_t1 + lambda1*openn_curr_par1_t1 + lambda2*openn_curr_par2_t1 + lambda3*openn_curr_par3_t1
openn_curr2 =~ NA*openn_curr_par1_t2 + lambda1*openn_curr_par1_t2 + lambda2*openn_curr_par2_t2 + lambda3*openn_curr_par3_t2
# Intercepts
openn_curr_par1_t1 ~ i1*1
openn_curr_par2_t1 ~ 1
openn_curr_par3_t1 ~ 1
openn_curr_par1_t2 ~ i1*1
openn_curr_par2_t2 ~ 1
openn_curr_par3_t2 ~ 1
# Unique Variances
openn_curr_par1_t1 ~~ openn_curr_par1_t1
openn_curr_par2_t1 ~~ openn_curr_par2_t1
openn_curr_par3_t1 ~~ openn_curr_par3_t1
openn_curr_par1_t2 ~~ openn_curr_par1_t2
openn_curr_par2_t2 ~~ openn_curr_par2_t2
openn_curr_par3_t2 ~~ openn_curr_par3_t2
# Latent Variable Means
openn_curr1 ~ 0*1
openn_curr2 ~ 1
# Latent Variable Variances and Covariance
openn_curr1 ~~ 1*openn_curr1
openn_curr2 ~~ openn_curr2
openn_curr1 ~~ openn_curr2
'
fit_weak_openn_curr <- cfa(weak_openn_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_weak_openn_curr, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_openn_curr <- '
# Define the latent factors
openn_curr1 =~ NA*openn_curr_par1_t1 + lambda1*openn_curr_par1_t1 + lambda2*openn_curr_par2_t1 + lambda3*openn_curr_par3_t1
openn_curr2 =~ NA*openn_curr_par1_t2 + lambda1*openn_curr_par1_t2 + lambda2*openn_curr_par2_t2 + lambda3*openn_curr_par3_t2
# Intercepts
openn_curr_par1_t1 ~ i1*1
openn_curr_par2_t1 ~ i2*1
openn_curr_par3_t1 ~ i3*1
openn_curr_par1_t2 ~ i1*1
openn_curr_par2_t2 ~ i2*1
openn_curr_par3_t2 ~ i3*1
# Unique Variances
openn_curr_par1_t1 ~~ openn_curr_par1_t1
openn_curr_par2_t1 ~~ openn_curr_par2_t1
openn_curr_par3_t1 ~~ openn_curr_par3_t1
openn_curr_par1_t2 ~~ openn_curr_par1_t2
openn_curr_par2_t2 ~~ openn_curr_par2_t2
openn_curr_par3_t2 ~~ openn_curr_par3_t2
# Latent Variable Means
openn_curr1 ~ 0*1
openn_curr2 ~ 1
# Latent Variable Variances and Covariance
openn_curr1 ~~ 1*openn_curr1
openn_curr2 ~~ openn_curr2
openn_curr1 ~~ openn_curr2
'
fit_strong_openn_curr <- cfa(strong_openn_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strong_openn_curr, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_openn_curr <- '
# Define the latent factors
openn_curr1 =~ NA*openn_curr_par1_t1 + lambda1*openn_curr_par1_t1 + lambda2*openn_curr_par2_t1 + lambda3*openn_curr_par3_t1
openn_curr2 =~ NA*openn_curr_par1_t2 + lambda1*openn_curr_par1_t2 + lambda2*openn_curr_par2_t2 + lambda3*openn_curr_par3_t2
# Intercepts
openn_curr_par1_t1 ~ i1*1
openn_curr_par2_t1 ~ i2*1
openn_curr_par3_t1 ~ i3*1
openn_curr_par1_t2 ~ i1*1
openn_curr_par2_t2 ~ i2*1
openn_curr_par3_t2 ~ i3*1
# Unique Variances
openn_curr_par1_t1 ~~ u1*openn_curr_par1_t1
openn_curr_par2_t1 ~~ u2*openn_curr_par2_t1
openn_curr_par3_t1 ~~ u3*openn_curr_par3_t1
openn_curr_par1_t2 ~~ u1*openn_curr_par1_t2
openn_curr_par2_t2 ~~ u2*openn_curr_par2_t2
openn_curr_par3_t2 ~~ u3*openn_curr_par3_t2
# Latent Variable Means
openn_curr1 ~ 0*1
openn_curr2 ~ 1
# Latent Variable Variances and Covariance
openn_curr1 ~~ 1*openn_curr1
openn_curr2 ~~ openn_curr2
openn_curr1 ~~ openn_curr2
'
fit_strict_openn_curr <- cfa(strict_openn_curr, data = df_sbsa3_wide_pers, mimic = "mplus", missing="ML")
summary(fit_strict_openn_curr, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_openn_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_openn_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_openn_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_openn_curr) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 530 19 365. 5357. 5438. 0.810 0.644 0.290 0.0789
2 weak 530 17 365. 5353. 5426. 0.811 0.716 0.259 0.0789
3 strong 530 15 368. 5352. 5416. 0.811 0.763 0.237 0.0799
4 strict 530 12 368. 5346. 5398. 0.812 0.812 0.211 0.0788# chi-square difference test for nested models
anova(fit_configural_openn_curr, fit_weak_openn_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_openn_curr 8 5356.7 5437.9 364.87
fit_weak_openn_curr 10 5352.9 5425.6 365.06 0.19134 0 2
Pr(>Chisq)
fit_configural_openn_curr
fit_weak_openn_curr 0.9088anova(fit_weak_openn_curr, fit_strong_openn_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_openn_curr 10 5352.9 5425.6 365.06
fit_strong_openn_curr 12 5351.6 5415.7 367.73 2.6715 0.025169 2
Pr(>Chisq)
fit_weak_openn_curr
fit_strong_openn_curr 0.263anova(fit_strong_openn_curr, fit_strict_openn_curr)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_openn_curr 12 5351.6 5415.7 367.73
fit_strict_openn_curr 15 5346.3 5397.6 368.41 0.68078 0 3
Pr(>Chisq)
fit_strong_openn_curr
fit_strict_openn_curr 0.8777Fit model:
# Code snippets adapted from https://quantdev.ssri.psu.edu/sites/qdev/files/LongitudinalMeasurementInvariance_2017_1108.html
# Configural invariance model
configural_openn_ideal <- '
# Define the latent factors
openn_ideal1 =~ NA*openn_ideal_par1_t1 + lambda1*openn_ideal_par1_t1 + openn_ideal_par2_t1 + openn_ideal_par3_t1
openn_ideal2 =~ NA*openn_ideal_par1_t2 + lambda1*openn_ideal_par1_t2 + openn_ideal_par2_t2 + openn_ideal_par3_t2
# Intercepts
openn_ideal_par1_t1 ~ i1*1
openn_ideal_par2_t1 ~ 1
openn_ideal_par3_t1 ~ 1
openn_ideal_par1_t2 ~ i1*1
openn_ideal_par2_t2 ~ 1
openn_ideal_par3_t2 ~ 1
# Unique Variances
openn_ideal_par1_t1 ~~ openn_ideal_par1_t1
openn_ideal_par2_t1 ~~ openn_ideal_par2_t1
openn_ideal_par3_t1 ~~ openn_ideal_par3_t1
openn_ideal_par1_t2 ~~ openn_ideal_par1_t2
openn_ideal_par2_t2 ~~ openn_ideal_par2_t2
openn_ideal_par3_t2 ~~ openn_ideal_par3_t2
# Latent Variable Means
openn_ideal1 ~ 0*1
openn_ideal2 ~ 1
# Latent Variable Variances and Covariance
openn_ideal1 ~~ 1*openn_ideal1
openn_ideal2 ~~ openn_ideal2
openn_ideal1 ~~ openn_ideal2
'
fit_configural_openn_ideal <- cfa(configural_openn_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_configural_openn_ideal, fit.measures = TRUE)
# Weak invariance model
weak_openn_ideal <- '
# Define the latent factors
openn_ideal1 =~ NA*openn_ideal_par1_t1 + lambda1*openn_ideal_par1_t1 + lambda2*openn_ideal_par2_t1 + lambda3*openn_ideal_par3_t1
openn_ideal2 =~ NA*openn_ideal_par1_t2 + lambda1*openn_ideal_par1_t2 + lambda2*openn_ideal_par2_t2 + lambda3*openn_ideal_par3_t2
# Intercepts
openn_ideal_par1_t1 ~ i1*1
openn_ideal_par2_t1 ~ 1
openn_ideal_par3_t1 ~ 1
openn_ideal_par1_t2 ~ i1*1
openn_ideal_par2_t2 ~ 1
openn_ideal_par3_t2 ~ 1
# Unique Variances
openn_ideal_par1_t1 ~~ openn_ideal_par1_t1
openn_ideal_par2_t1 ~~ openn_ideal_par2_t1
openn_ideal_par3_t1 ~~ openn_ideal_par3_t1
openn_ideal_par1_t2 ~~ openn_ideal_par1_t2
openn_ideal_par2_t2 ~~ openn_ideal_par2_t2
openn_ideal_par3_t2 ~~ openn_ideal_par3_t2
# Latent Variable Means
openn_ideal1 ~ 0*1
openn_ideal2 ~ 1
# Latent Variable Variances and Covariance
openn_ideal1 ~~ 1*openn_ideal1
openn_ideal2 ~~ openn_ideal2
openn_ideal1 ~~ openn_ideal2
'
fit_weak_openn_ideal <- cfa(weak_openn_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_weak_openn_ideal, fit.measures = TRUE)
# Strong invariance model (additional constraints on manifest intercepts)
strong_openn_ideal <- '
# Define the latent factors
openn_ideal1 =~ NA*openn_ideal_par1_t1 + lambda1*openn_ideal_par1_t1 + lambda2*openn_ideal_par2_t1 + lambda3*openn_ideal_par3_t1
openn_ideal2 =~ NA*openn_ideal_par1_t2 + lambda1*openn_ideal_par1_t2 + lambda2*openn_ideal_par2_t2 + lambda3*openn_ideal_par3_t2
# Intercepts
openn_ideal_par1_t1 ~ i1*1
openn_ideal_par2_t1 ~ i2*1
openn_ideal_par3_t1 ~ i3*1
openn_ideal_par1_t2 ~ i1*1
openn_ideal_par2_t2 ~ i2*1
openn_ideal_par3_t2 ~ i3*1
# Unique Variances
openn_ideal_par1_t1 ~~ openn_ideal_par1_t1
openn_ideal_par2_t1 ~~ openn_ideal_par2_t1
openn_ideal_par3_t1 ~~ openn_ideal_par3_t1
openn_ideal_par1_t2 ~~ openn_ideal_par1_t2
openn_ideal_par2_t2 ~~ openn_ideal_par2_t2
openn_ideal_par3_t2 ~~ openn_ideal_par3_t2
# Latent Variable Means
openn_ideal1 ~ 0*1
openn_ideal2 ~ 1
# Latent Variable Variances and Covariance
openn_ideal1 ~~ 1*openn_ideal1
openn_ideal2 ~~ openn_ideal2
openn_ideal1 ~~ openn_ideal2
'
fit_strong_openn_ideal <- cfa(strong_openn_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_strong_openn_ideal, fit.measures = TRUE)
# Strict invariance model (additional constraints on manifest variances)
strict_openn_ideal <- '
# Define the latent factors
openn_ideal1 =~ NA*openn_ideal_par1_t1 + lambda1*openn_ideal_par1_t1 + lambda2*openn_ideal_par2_t1 + lambda3*openn_ideal_par3_t1
openn_ideal2 =~ NA*openn_ideal_par1_t2 + lambda1*openn_ideal_par1_t2 + lambda2*openn_ideal_par2_t2 + lambda3*openn_ideal_par3_t2
# Intercepts
openn_ideal_par1_t1 ~ i1*1
openn_ideal_par2_t1 ~ i2*1
openn_ideal_par3_t1 ~ i3*1
openn_ideal_par1_t2 ~ i1*1
openn_ideal_par2_t2 ~ i2*1
openn_ideal_par3_t2 ~ i3*1
# Unique Variances
openn_ideal_par1_t1 ~~ u1*openn_ideal_par1_t1
openn_ideal_par2_t1 ~~ u2*openn_ideal_par2_t1
openn_ideal_par3_t1 ~~ u3*openn_ideal_par3_t1
openn_ideal_par1_t2 ~~ u1*openn_ideal_par1_t2
openn_ideal_par2_t2 ~~ u2*openn_ideal_par2_t2
openn_ideal_par3_t2 ~~ u3*openn_ideal_par3_t2
# Latent Variable Means
openn_ideal1 ~ 0*1
openn_ideal2 ~ 1
# Latent Variable Variances and Covariance
openn_ideal1 ~~ 1*openn_ideal1
openn_ideal2 ~~ openn_ideal2
openn_ideal1 ~~ openn_ideal2
'
fit_strict_openn_ideal <- cfa(strict_openn_ideal, data = df_sbsa3_wide_pers %>% filter(group=="Group 1"), mimic = "mplus", missing="ML")
summary(fit_strict_openn_ideal, fit.measures = TRUE)Results summary:
# compare model fit
bind_rows(broom::glance(fit_configural_openn_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_weak_openn_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strong_openn_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr),
broom::glance(fit_strict_openn_ideal) %>%
select(nobs, npar, chisq, AIC, BIC, cfi, tli, rmsea, srmr)) %>%
mutate(model = c("configural", "weak", "strong", "strict")) %>%
select(model, everything())# A tibble: 4 × 10
model nobs npar chisq AIC BIC cfi tli rmsea srmr
<chr> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 configural 175 19 35.9 1696. 1756. 0.922 0.853 0.141 0.0487
2 weak 175 17 37.6 1693. 1747. 0.923 0.884 0.126 0.0705
3 strong 175 15 38.2 1690. 1738. 0.927 0.908 0.112 0.0658
4 strict 175 12 40.4 1686. 1724. 0.929 0.929 0.0984 0.0786# chi-square difference test for nested models
anova(fit_configural_openn_ideal, fit_weak_openn_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_configural_openn_ideal 8 1695.8 1755.9 35.920
fit_weak_openn_ideal 10 1693.4 1747.2 37.597 1.6775 0 2
Pr(>Chisq)
fit_configural_openn_ideal
fit_weak_openn_ideal 0.4323anova(fit_weak_openn_ideal, fit_strong_openn_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_weak_openn_ideal 10 1693.4 1747.2 37.597
fit_strong_openn_ideal 12 1690.0 1737.5 38.211 0.61394 0 2
Pr(>Chisq)
fit_weak_openn_ideal
fit_strong_openn_ideal 0.7357anova(fit_strong_openn_ideal, fit_strict_openn_ideal)
Chi-Squared Difference Test
Df AIC BIC Chisq Chisq diff RMSEA Df diff
fit_strong_openn_ideal 12 1690.0 1737.5 38.211
fit_strict_openn_ideal 15 1686.2 1724.2 40.411 2.1997 0 3
Pr(>Chisq)
fit_strong_openn_ideal
fit_strict_openn_ideal 0.532How many participants at each time point and in each condition?
df_sbsa3 %>% summarise(n_distinct(pid)) # N (after exclusions)# A tibble: 1 × 1
`n_distinct(pid)`
<int>
1 530df_sbsa3 %>% group_by(time, group) %>% tally()# A tibble: 6 × 3
# Groups: time [2]
time group n
<dbl> <chr> <int>
1 1 Group 1 174
2 1 Group 2 179
3 1 Group 3 176
4 2 Group 1 155
5 2 Group 2 152
6 2 Group 3 159One in Group 1 only has T2 data.
Gender and age distribution
# across all participants
df_sbsa3 %>% group_by(pid) %>% slice_head(n=1) %>% ungroup() %>%
mutate(gender_num = ifelse(gender=="Female", 1, 0)) %>%
summarise(n = n(), women = sum(gender_num), m_age = mean(age, na.rm=T), sd_age = sd(age, na.rm=T), min_age = min(age, na.rm=T), max_age = max(age, na.rm=T)) %>% kable(digits = 2)| n | women | m_age | sd_age | min_age | max_age |
|---|---|---|---|---|---|
| 530 | 265 | 31.17 | 10.11 | 19 | 105 |
# by group and time point
df_sbsa3 %>% group_by(time, group) %>%
summarise(n = n(), m_age = mean(age, na.rm=T), sd_age = sd(age, na.rm=T), min_age = min(age, na.rm=T), max_age = max(age, na.rm=T)) %>% kable(digits = 2)| time | group | n | m_age | sd_age | min_age | max_age |
|---|---|---|---|---|---|---|
| 1 | Group 1 | 174 | 31.46 | 9.54 | 19 | 66 |
| 1 | Group 2 | 179 | 31.12 | 10.58 | 20 | 105 |
| 1 | Group 3 | 176 | 30.93 | 10.26 | 19 | 70 |
| 2 | Group 1 | 155 | 31.75 | 9.52 | 20 | 66 |
| 2 | Group 2 | 152 | 31.84 | 10.99 | 20 | 105 |
| 2 | Group 3 | 159 | 30.70 | 10.24 | 19 | 70 |
df_sbsa3 %>% group_by(time, group, gender) %>%
summarise(n = n(), m_age = mean(age, na.rm=T), sd_age = sd(age, na.rm=T), min_age = min(age, na.rm=T), max_age = max(age, na.rm=T)) %>% kable(digits = 2)| time | group | gender | n | m_age | sd_age | min_age | max_age |
|---|---|---|---|---|---|---|---|
| 1 | Group 1 | Female | 88 | 32.11 | 10.17 | 21 | 66 |
| 1 | Group 1 | Male | 86 | 30.79 | 8.86 | 19 | 66 |
| 1 | Group 2 | Female | 90 | 31.73 | 11.26 | 20 | 105 |
| 1 | Group 2 | Male | 89 | 30.52 | 9.87 | 20 | 62 |
| 1 | Group 3 | Female | 87 | 29.91 | 9.82 | 21 | 66 |
| 1 | Group 3 | Male | 89 | 31.92 | 10.64 | 19 | 70 |
| 2 | Group 1 | Female | 81 | 32.20 | 10.09 | 21 | 66 |
| 2 | Group 1 | Male | 74 | 31.26 | 8.90 | 20 | 66 |
| 2 | Group 2 | Female | 81 | 31.88 | 11.52 | 20 | 105 |
| 2 | Group 2 | Male | 71 | 31.80 | 10.45 | 20 | 62 |
| 2 | Group 3 | Female | 81 | 29.83 | 9.91 | 21 | 66 |
| 2 | Group 3 | Male | 78 | 31.62 | 10.56 | 19 | 70 |
Other demographics (from Prolific info):
# across all participants: ethnicity
df_sbsa3 %>% group_by(pid) %>% slice_head(n=1) %>% ungroup() %>%
group_by(ethnicity) %>% summarise(n = n()) %>% ungroup() %>% mutate(prop = round(n / sum(n) * 100, 3)) %>%
kable(digits = 1)| ethnicity | n | prop |
|---|---|---|
| Asian | 17 | 3.2 |
| Black | 116 | 21.9 |
| Mixed | 35 | 6.6 |
| Other | 23 | 4.3 |
| White | 336 | 63.4 |
| NA | 3 | 0.6 |
# across all participants: country of residence
df_sbsa3 %>% group_by(pid) %>% slice_head(n=1) %>% ungroup() %>%
group_by(country) %>% summarise(n = n()) %>% ungroup() %>%
arrange(desc(n)) %>% mutate(prop = round(n / sum(n) * 100, 3)) %>% print(n=20) %>% kable(digits = 1)# A tibble: 32 × 3
country n prop
<chr> <int> <dbl>
1 South Africa 138 26.0
2 Portugal 89 16.8
3 Poland 69 13.0
4 Italy 41 7.74
5 Mexico 30 5.66
6 United Kingdom 24 4.53
7 Greece 19 3.58
8 Spain 17 3.21
9 Chile 16 3.02
10 Hungary 13 2.45
11 Canada 8 1.51
12 Australia 6 1.13
13 France 6 1.13
14 Germany 6 1.13
15 Ireland 6 1.13
16 Finland 5 0.943
17 Latvia 5 0.943
18 Netherlands 5 0.943
19 Belgium 3 0.566
20 Israel 3 0.566
# ℹ 12 more rows| country | n | prop |
|---|---|---|
| South Africa | 138 | 26.0 |
| Portugal | 89 | 16.8 |
| Poland | 69 | 13.0 |
| Italy | 41 | 7.7 |
| Mexico | 30 | 5.7 |
| United Kingdom | 24 | 4.5 |
| Greece | 19 | 3.6 |
| Spain | 17 | 3.2 |
| Chile | 16 | 3.0 |
| Hungary | 13 | 2.5 |
| Canada | 8 | 1.5 |
| Australia | 6 | 1.1 |
| France | 6 | 1.1 |
| Germany | 6 | 1.1 |
| Ireland | 6 | 1.1 |
| Finland | 5 | 0.9 |
| Latvia | 5 | 0.9 |
| Netherlands | 5 | 0.9 |
| Belgium | 3 | 0.6 |
| Israel | 3 | 0.6 |
| Slovenia | 3 | 0.6 |
| Switzerland | 3 | 0.6 |
| Austria | 2 | 0.4 |
| Czech Republic | 2 | 0.4 |
| Estonia | 2 | 0.4 |
| Norway | 2 | 0.4 |
| Sweden | 2 | 0.4 |
| Iceland | 1 | 0.2 |
| India | 1 | 0.2 |
| Korea | 1 | 0.2 |
| New Zealand | 1 | 0.2 |
| NA | 1 | 0.2 |
# across all participants: student status
df_sbsa3 %>% group_by(pid) %>% slice_head(n=1) %>% ungroup() %>%
group_by(student) %>% summarise(n = n()) %>% ungroup() %>%
arrange(desc(n)) %>% mutate(prop = round(n / sum(n) * 100, 3)) %>% kable(digits = 1)| student | n | prop |
|---|---|---|
| No | 280 | 52.8 |
| Yes | 163 | 30.8 |
| NA | 87 | 16.4 |
# across all participants: employment status
df_sbsa3 %>% group_by(pid) %>% slice_head(n=1) %>% ungroup() %>%
group_by(employed) %>% summarise(n = n()) %>% ungroup() %>%
arrange(desc(n)) %>% mutate(prop = round(n / sum(n) * 100, 3)) %>% kable(digits = 1)| employed | n | prop |
|---|---|---|
| Full-Time | 213 | 40.2 |
| NA | 96 | 18.1 |
| Unemployed (and job seeking) | 87 | 16.4 |
| Part-Time | 68 | 12.8 |
| Other | 33 | 6.2 |
| Not in paid work (e.g. homemaker’, ’retired or disabled) | 22 | 4.2 |
| Due to start a new job within the next month | 11 | 2.1 |
On the trait/dimensional level:
# created in study 3 cleaning script
base::load("data/int_consist_traits_st3.rda")
kable(int_consist_traits_st3, digits = 3)| trait | rel_alpha_current | rel_alpha_ideal | rel_omega_t_current | rel_omega_t_ideal | rel_omega_h_current | rel_omega_h_ideal |
|---|---|---|---|---|---|---|
| lifesat | 0.874 | NA | 0.895 | NA | 0.828 | NA |
| meaning | 0.867 | NA | 0.948 | NA | 0.580 | NA |
| selfes | 0.909 | NA | 0.936 | NA | 0.833 | NA |
| concept | 0.893 | NA | 0.914 | NA | 0.860 | NA |
| extra | 0.862 | 0.730 | 0.893 | 0.773 | 0.688 | 0.480 |
| agree | 0.789 | 0.741 | 0.842 | 0.806 | 0.584 | 0.595 |
| consc | 0.871 | 0.732 | 0.896 | 0.796 | 0.683 | 0.579 |
| neuro | 0.913 | 0.749 | 0.928 | 0.803 | 0.757 | 0.562 |
| openn | 0.814 | 0.745 | 0.856 | 0.808 | 0.691 | 0.602 |
McDonald’s omega only really makes sense for the Big Five dimensions (with their three facets each) but I’ve added it for the well-being measures as well. Keep in mind that a few of the omega computations for the well-being measures only converged with error messages. These constructs were not neccessarily intended to be tested in this multidimensional manner for internal consistency.
On the facet level:
# created in study 3 cleaning script
base::load("data/int_consist_facets_st3.rda")
kable(int_consist_facets_st3, digits = 3)| facet | rel_alpha_current | rel_alpha_ideal |
|---|---|---|
| socia | 0.830 | 0.621 |
| asser | 0.748 | 0.546 |
| energ | 0.737 | 0.423 |
| compa | 0.544 | 0.386 |
| respe | 0.686 | 0.616 |
| trust | 0.677 | 0.612 |
| organ | 0.835 | 0.519 |
| produ | 0.778 | 0.622 |
| respo | 0.635 | 0.407 |
| anxie | 0.791 | 0.402 |
| depre | 0.836 | 0.508 |
| volat | 0.827 | 0.578 |
| curio | 0.635 | 0.534 |
| aesth | 0.646 | 0.415 |
| imagi | 0.730 | 0.618 |
Do participants who only took part in T1 differ from those taking part in T1 and T2?
Determine groups of “remainers” vs. “drop-outs” and run analyses:
attrition_sample <- df_sbsa3 %>%
group_by(pid) %>% mutate(last_time = max(time)) %>% ungroup() %>%
filter(time==1) %>%
select(pid, last_time,
contains("_comb_"), c("swls", "meaning", "selfes", "concept"), # outcomes
gender, age, ethnicity, student, employed) %>% # demographics
mutate(remained = last_time - 1,
female = if_else(gender=="Female", 1, 0, NA_real_),
white = if_else(ethnicity=="White", 1, 0, NA_real_),
black = if_else(ethnicity=="Black", 1, 0, NA_real_),
student = if_else(student=="Yes", 1, 0, NA_real_),
fulltime = if_else(employed=="Full-Time", 1, 0, NA_real_),
parttime = if_else(employed=="Part-Time", 1, 0, NA_real_),
notemployed = if_else(employed=="Unemployed (and job seeking)" |
employed=="Not in paid work (e.g. homemaker', 'retired or disabled)" |
employed=="Due to start a new job within the next month", 1, 0, NA_real_),) %>%
select(-c(last_time, na_comb_curr, na_comb_ideal, gender, ethnicity, employed))
attrition_analysis <- colnames(attrition_sample) %>% # Start with all column names
setdiff( c("pid", "remained" ) ) %>% # ...that are not Sepal.Length
rlang::syms() %>% # Convert them to symbols
map( ~rlang::expr(lm(!!.x ~ remained,
data=attrition_sample)) ) %>% # Create expressions
map( eval.parent ) %>% # Evaluate expressions
map( broom::tidy ) %>% # Tidy up the output
bind_rows() %>% # Combine into a single data frame
filter( term != "(Intercept)" ) %>% # Drop all (Intercept) entries
mutate(outcome = colnames(attrition_sample) %>% setdiff( c("pid", "remained" ) ))Display significant differences:
attrition_analysis %>% filter(p.value < .05) %>% arrange(desc(abs(estimate)))# A tibble: 4 × 6
term estimate std.error statistic p.value outcome
<chr> <dbl> <dbl> <dbl> <dbl> <chr>
1 remained 0.230 0.0957 2.41 0.0164 compa_comb_curr
2 remained 0.179 0.0663 2.70 0.00724 female
3 remained 0.142 0.0549 2.59 0.00985 black
4 remained -0.127 0.0640 -1.98 0.0482 white effectsize::cohens_d(compa_comb_curr ~ remained, data = attrition_sample)Cohen's d | 95% CI
--------------------------
-0.32 | [-0.58, -0.06]
- Estimated using pooled SD.effectsize::phi(table(attrition_sample$female, attrition_sample$remained), alternative = "two.sided")Phi (adj.) | 95% CI
-------------------------
0.11 | [0.00, 0.20]effectsize::phi(table(attrition_sample$black, attrition_sample$remained), alternative = "two.sided")Phi (adj.) | 95% CI
-------------------------
0.10 | [0.00, 0.19]effectsize::phi(table(attrition_sample$white, attrition_sample$remained), alternative = "two.sided")Phi (adj.) | 95% CI
-------------------------
0.07 | [0.00, 0.17]Participants who dropped out at T2 were less compassionate, d = -0.32, 95% CI [-0.58; -0.06], p = .016, less likely to be female, phi = .11, 95% CI [0; .20], p = .007, and Black, phi = .10, 95% CI [0; .19], p = .001, and more likely to be White, phi = .07, 95% CI [0; .17], p = .048.
Show all results:
attrition_analysis %>% arrange(desc(abs(estimate))) %>% print(n=200)# A tibble: 52 × 6
term estimate std.error statistic p.value outcome
<chr> <dbl> <dbl> <dbl> <dbl> <chr>
1 remained 2.11 1.35 1.57 0.118 age
2 remained 0.255 0.152 1.68 0.0932 meaning
3 remained 0.230 0.0957 2.41 0.0164 compa_comb_curr
4 remained -0.212 0.146 -1.46 0.147 anxie_comb_ideal
5 remained 0.193 0.134 1.44 0.152 produ_comb_ideal
6 remained 0.192 0.162 1.19 0.237 compa_comb_ideal
7 remained 0.191 0.161 1.18 0.240 asser_comb_ideal
8 remained -0.186 0.142 -1.31 0.192 volat_comb_ideal
9 remained 0.179 0.0663 2.70 0.00724 female
10 remained -0.177 0.113 -1.57 0.119 neuro_comb_ideal
11 remained 0.172 0.162 1.06 0.290 curio_comb_ideal
12 remained 0.171 0.112 1.53 0.127 aesth_comb_curr
13 remained 0.164 0.109 1.50 0.135 consc_comb_ideal
14 remained 0.160 0.140 1.14 0.255 organ_comb_ideal
15 remained 0.159 0.117 1.37 0.173 concept
16 remained -0.155 0.127 -1.22 0.222 anxie_comb_curr
17 remained -0.148 0.138 -1.07 0.285 socia_comb_curr
18 remained 0.146 0.0784 1.86 0.0628 agree_comb_curr
19 remained 0.142 0.0549 2.59 0.00985 black
20 remained 0.140 0.119 1.17 0.241 produ_comb_curr
21 remained 0.137 0.139 0.985 0.326 respo_comb_ideal
22 remained 0.134 0.131 1.02 0.307 openn_comb_ideal
23 remained -0.134 0.137 -0.973 0.331 volat_comb_curr
24 remained 0.133 0.0855 1.56 0.120 openn_comb_curr
25 remained -0.133 0.128 -1.04 0.301 depre_comb_ideal
26 remained -0.130 0.0666 -1.96 0.0509 notemployed
27 remained 0.129 0.172 0.750 0.454 imagi_comb_ideal
28 remained -0.129 0.178 -0.728 0.468 socia_comb_ideal
29 remained -0.127 0.0640 -1.98 0.0482 white
30 remained 0.125 0.0919 1.36 0.175 respe_comb_curr
31 remained 0.119 0.109 1.09 0.277 imagi_comb_curr
32 remained -0.117 0.118 -0.987 0.324 neuro_comb_curr
33 remained -0.110 0.120 -0.917 0.359 asser_comb_curr
34 remained 0.109 0.0958 1.14 0.254 curio_comb_curr
35 remained 0.102 0.164 0.618 0.537 aesth_comb_ideal
36 remained 0.101 0.134 0.751 0.453 swls
37 remained 0.0853 0.121 0.703 0.483 agree_comb_ideal
38 remained 0.0834 0.109 0.764 0.445 trust_comb_curr
39 remained -0.0672 0.101 -0.662 0.508 extra_comb_curr
40 remained -0.0617 0.136 -0.454 0.650 depre_comb_curr
41 remained 0.0601 0.118 0.508 0.612 selfes
42 remained 0.0583 0.129 0.451 0.653 respe_comb_ideal
43 remained 0.0556 0.116 0.478 0.633 energ_comb_curr
44 remained 0.0515 0.101 0.511 0.610 consc_comb_curr
45 remained 0.0447 0.0543 0.822 0.412 parttime
46 remained 0.0438 0.0747 0.586 0.558 fulltime
47 remained 0.0365 0.143 0.256 0.798 energ_comb_ideal
48 remained 0.0326 0.125 0.261 0.795 extra_comb_ideal
49 remained 0.0143 0.101 0.141 0.888 respo_comb_curr
50 remained -0.00952 0.0738 -0.129 0.897 student
51 remained 0.00552 0.174 0.0317 0.975 trust_comb_ideal
52 remained 0.000109 0.136 0.000802 0.999 organ_comb_curr Prepare data frame for plotting:
df_manip_check_forplot <- df_sbsa3 %>%
mutate(time_d = time - 1) %>%
select(pid, time_d, group, contains("_comb_"), -c(na_comb_curr, na_comb_ideal)) %>%
pivot_longer(-c(pid, time_d, group),
cols_vary = "slowest",
names_to = c("trait", "reference"),
names_pattern = "(.*)_(.*)",
values_to = "score") %>%
mutate(trait = gsub("_comb", "", trait),
reference = ifelse(reference=="curr", "current", reference)) %>%
left_join(tibble(long = str_to_title(names(b5_vars)), trait = str_trunc(names(b5_vars), 5, ellipsis = ""))) %>%
select(-trait, trait = long, time = time_d) %>%
filter(!is.na(score)) %>%
group_by(trait, reference, group, time) %>%
summarise(n = n(), m_trait = mean(score), sd_trait = sd(score)) %>%
ungroup() %>%
mutate(Group = factor(group, levels = c("Group 1", "Group 2", "Group 3"), labels = c("Group 1", "Group 2", "Group 3")),
Time = factor(time+1),
Trait = factor(trait,
levels = str_to_title(names(b5_vars))[c(1,6:8, 2,9:11, 3,12:14, 4,15:17, 5,18:20)], # group facets with each trait
labels = str_to_title(names(b5_vars))[c(1,6:8, 2,9:11, 3,12:14, 4,15:17, 5,18:20)]),
Reference = factor(reference, levels = c("current", "ideal"), labels = c("Current Personality", "Ideal Personality")),
lwr95 = m_trait - qnorm(0.975)*sd_trait/sqrt(n),
upr95 = m_trait + qnorm(0.975)*sd_trait/sqrt(n)) %>%
select(-c(time, group, trait, reference))Table of all outcomes:
df_manip_check_forplot %>% arrange(Trait) %>%
mutate_if(is.numeric, round, digits=2) %>%
select(Trait, Reference, Group, Time, n, m_trait, lwr95, upr95, sd_trait) %>%
kable(digits = 2)| Trait | Reference | Group | Time | n | m_trait | lwr95 | upr95 | sd_trait |
|---|---|---|---|---|---|---|---|---|
| Extraversion | Current Personality | Group 1 | 1 | 174 | 2.83 | 2.72 | 2.94 | 0.75 |
| Extraversion | Current Personality | Group 1 | 2 | 155 | 2.89 | 2.77 | 3.01 | 0.75 |
| Extraversion | Current Personality | Group 2 | 1 | 179 | 2.92 | 2.81 | 3.03 | 0.76 |
| Extraversion | Current Personality | Group 2 | 2 | 152 | 3.00 | 2.89 | 3.12 | 0.74 |
| Extraversion | Current Personality | Group 3 | 1 | 176 | 3.07 | 2.96 | 3.18 | 0.76 |
| Extraversion | Current Personality | Group 3 | 2 | 159 | 3.03 | 2.91 | 3.15 | 0.80 |
| Extraversion | Ideal Personality | Group 1 | 1 | 174 | 3.92 | 3.84 | 4.00 | 0.52 |
| Extraversion | Ideal Personality | Group 1 | 2 | 155 | 3.82 | 3.74 | 3.90 | 0.51 |
| Sociability | Current Personality | Group 1 | 1 | 174 | 2.46 | 2.30 | 2.61 | 1.04 |
| Sociability | Current Personality | Group 1 | 2 | 155 | 2.57 | 2.41 | 2.73 | 1.01 |
| Sociability | Current Personality | Group 2 | 1 | 179 | 2.63 | 2.49 | 2.78 | 1.00 |
| Sociability | Current Personality | Group 2 | 2 | 152 | 2.75 | 2.61 | 2.90 | 0.93 |
| Sociability | Current Personality | Group 3 | 1 | 176 | 2.70 | 2.54 | 2.85 | 1.06 |
| Sociability | Current Personality | Group 3 | 2 | 159 | 2.70 | 2.53 | 2.86 | 1.07 |
| Sociability | Ideal Personality | Group 1 | 1 | 174 | 3.65 | 3.54 | 3.76 | 0.75 |
| Sociability | Ideal Personality | Group 1 | 2 | 155 | 3.51 | 3.40 | 3.63 | 0.73 |
| Assertiveness | Current Personality | Group 1 | 1 | 174 | 2.92 | 2.79 | 3.05 | 0.89 |
| Assertiveness | Current Personality | Group 1 | 2 | 155 | 2.92 | 2.78 | 3.06 | 0.89 |
| Assertiveness | Current Personality | Group 2 | 1 | 179 | 2.97 | 2.83 | 3.10 | 0.91 |
| Assertiveness | Current Personality | Group 2 | 2 | 152 | 2.99 | 2.85 | 3.12 | 0.87 |
| Assertiveness | Current Personality | Group 3 | 1 | 176 | 3.17 | 3.05 | 3.30 | 0.87 |
| Assertiveness | Current Personality | Group 3 | 2 | 159 | 3.11 | 2.96 | 3.26 | 0.95 |
| Assertiveness | Ideal Personality | Group 1 | 1 | 174 | 3.89 | 3.79 | 3.99 | 0.68 |
| Assertiveness | Ideal Personality | Group 1 | 2 | 155 | 3.75 | 3.64 | 3.86 | 0.67 |
| Energy | Current Personality | Group 1 | 1 | 174 | 3.12 | 2.99 | 3.24 | 0.84 |
| Energy | Current Personality | Group 1 | 2 | 155 | 3.19 | 3.05 | 3.32 | 0.87 |
| Energy | Current Personality | Group 2 | 1 | 179 | 3.16 | 3.03 | 3.29 | 0.87 |
| Energy | Current Personality | Group 2 | 2 | 152 | 3.27 | 3.14 | 3.41 | 0.87 |
| Energy | Current Personality | Group 3 | 1 | 176 | 3.34 | 3.20 | 3.47 | 0.89 |
| Energy | Current Personality | Group 3 | 2 | 159 | 3.28 | 3.14 | 3.43 | 0.93 |
| Energy | Ideal Personality | Group 1 | 1 | 174 | 4.22 | 4.13 | 4.31 | 0.60 |
| Energy | Ideal Personality | Group 1 | 2 | 155 | 4.19 | 4.10 | 4.28 | 0.57 |
| Agreeableness | Current Personality | Group 1 | 1 | 174 | 3.69 | 3.61 | 3.77 | 0.56 |
| Agreeableness | Current Personality | Group 1 | 2 | 155 | 3.76 | 3.67 | 3.85 | 0.55 |
| Agreeableness | Current Personality | Group 2 | 1 | 179 | 3.62 | 3.53 | 3.70 | 0.59 |
| Agreeableness | Current Personality | Group 2 | 2 | 152 | 3.68 | 3.59 | 3.78 | 0.60 |
| Agreeableness | Current Personality | Group 3 | 1 | 176 | 3.69 | 3.60 | 3.78 | 0.62 |
| Agreeableness | Current Personality | Group 3 | 2 | 159 | 3.68 | 3.58 | 3.78 | 0.64 |
| Agreeableness | Ideal Personality | Group 1 | 1 | 174 | 4.07 | 4.00 | 4.15 | 0.51 |
| Agreeableness | Ideal Personality | Group 1 | 2 | 155 | 4.05 | 3.97 | 4.14 | 0.54 |
| Compassion | Current Personality | Group 1 | 1 | 174 | 3.80 | 3.70 | 3.90 | 0.67 |
| Compassion | Current Personality | Group 1 | 2 | 155 | 3.94 | 3.84 | 4.05 | 0.67 |
| Compassion | Current Personality | Group 2 | 1 | 179 | 3.78 | 3.67 | 3.89 | 0.72 |
| Compassion | Current Personality | Group 2 | 2 | 152 | 3.78 | 3.66 | 3.90 | 0.76 |
| Compassion | Current Personality | Group 3 | 1 | 176 | 3.81 | 3.69 | 3.92 | 0.77 |
| Compassion | Current Personality | Group 3 | 2 | 159 | 3.80 | 3.68 | 3.93 | 0.80 |
| Compassion | Ideal Personality | Group 1 | 1 | 174 | 4.02 | 3.92 | 4.12 | 0.68 |
| Compassion | Ideal Personality | Group 1 | 2 | 155 | 4.05 | 3.94 | 4.16 | 0.69 |
| Respectfulness | Current Personality | Group 1 | 1 | 174 | 4.06 | 3.96 | 4.17 | 0.71 |
| Respectfulness | Current Personality | Group 1 | 2 | 155 | 4.09 | 3.98 | 4.21 | 0.71 |
| Respectfulness | Current Personality | Group 2 | 1 | 179 | 3.96 | 3.86 | 4.05 | 0.65 |
| Respectfulness | Current Personality | Group 2 | 2 | 152 | 4.03 | 3.93 | 4.13 | 0.64 |
| Respectfulness | Current Personality | Group 3 | 1 | 176 | 4.01 | 3.90 | 4.11 | 0.71 |
| Respectfulness | Current Personality | Group 3 | 2 | 159 | 4.04 | 3.92 | 4.15 | 0.72 |
| Respectfulness | Ideal Personality | Group 1 | 1 | 174 | 4.46 | 4.38 | 4.54 | 0.54 |
| Respectfulness | Ideal Personality | Group 1 | 2 | 155 | 4.41 | 4.32 | 4.51 | 0.60 |
| Trust | Current Personality | Group 1 | 1 | 174 | 3.21 | 3.09 | 3.33 | 0.81 |
| Trust | Current Personality | Group 1 | 2 | 155 | 3.25 | 3.13 | 3.37 | 0.77 |
| Trust | Current Personality | Group 2 | 1 | 179 | 3.11 | 2.99 | 3.23 | 0.80 |
| Trust | Current Personality | Group 2 | 2 | 152 | 3.24 | 3.11 | 3.37 | 0.81 |
| Trust | Current Personality | Group 3 | 1 | 176 | 3.25 | 3.13 | 3.38 | 0.84 |
| Trust | Current Personality | Group 3 | 2 | 159 | 3.21 | 3.07 | 3.34 | 0.87 |
| Trust | Ideal Personality | Group 1 | 1 | 174 | 3.73 | 3.62 | 3.84 | 0.73 |
| Trust | Ideal Personality | Group 1 | 2 | 155 | 3.69 | 3.58 | 3.81 | 0.72 |
| Conscientiousness | Current Personality | Group 1 | 1 | 174 | 3.33 | 3.22 | 3.44 | 0.76 |
| Conscientiousness | Current Personality | Group 1 | 2 | 155 | 3.42 | 3.31 | 3.54 | 0.74 |
| Conscientiousness | Current Personality | Group 2 | 1 | 179 | 3.36 | 3.26 | 3.47 | 0.72 |
| Conscientiousness | Current Personality | Group 2 | 2 | 152 | 3.45 | 3.34 | 3.56 | 0.69 |
| Conscientiousness | Current Personality | Group 3 | 1 | 176 | 3.54 | 3.43 | 3.66 | 0.77 |
| Conscientiousness | Current Personality | Group 3 | 2 | 159 | 3.58 | 3.46 | 3.70 | 0.76 |
| Conscientiousness | Ideal Personality | Group 1 | 1 | 174 | 4.39 | 4.33 | 4.46 | 0.46 |
| Conscientiousness | Ideal Personality | Group 1 | 2 | 155 | 4.41 | 4.34 | 4.47 | 0.42 |
| Organization | Current Personality | Group 1 | 1 | 174 | 3.41 | 3.25 | 3.56 | 1.05 |
| Organization | Current Personality | Group 1 | 2 | 155 | 3.48 | 3.32 | 3.64 | 1.00 |
| Organization | Current Personality | Group 2 | 1 | 179 | 3.45 | 3.30 | 3.59 | 1.00 |
| Organization | Current Personality | Group 2 | 2 | 152 | 3.50 | 3.35 | 3.65 | 0.95 |
| Organization | Current Personality | Group 3 | 1 | 176 | 3.68 | 3.54 | 3.83 | 0.99 |
| Organization | Current Personality | Group 3 | 2 | 159 | 3.63 | 3.47 | 3.78 | 0.98 |
| Organization | Ideal Personality | Group 1 | 1 | 174 | 4.42 | 4.33 | 4.50 | 0.59 |
| Organization | Ideal Personality | Group 1 | 2 | 155 | 4.43 | 4.35 | 4.51 | 0.52 |
| Productiveness | Current Personality | Group 1 | 1 | 174 | 3.03 | 2.89 | 3.16 | 0.93 |
| Productiveness | Current Personality | Group 1 | 2 | 155 | 3.17 | 3.03 | 3.31 | 0.90 |
| Productiveness | Current Personality | Group 2 | 1 | 179 | 3.16 | 3.04 | 3.28 | 0.82 |
| Productiveness | Current Personality | Group 2 | 2 | 152 | 3.29 | 3.16 | 3.43 | 0.84 |
| Productiveness | Current Personality | Group 3 | 1 | 176 | 3.35 | 3.21 | 3.48 | 0.91 |
| Productiveness | Current Personality | Group 3 | 2 | 159 | 3.49 | 3.35 | 3.63 | 0.90 |
| Productiveness | Ideal Personality | Group 1 | 1 | 174 | 4.51 | 4.42 | 4.59 | 0.57 |
| Productiveness | Ideal Personality | Group 1 | 2 | 155 | 4.50 | 4.42 | 4.58 | 0.51 |
| Responsibility | Current Personality | Group 1 | 1 | 174 | 3.56 | 3.45 | 3.67 | 0.77 |
| Responsibility | Current Personality | Group 1 | 2 | 155 | 3.62 | 3.50 | 3.74 | 0.78 |
| Responsibility | Current Personality | Group 2 | 1 | 179 | 3.49 | 3.37 | 3.60 | 0.77 |
| Responsibility | Current Personality | Group 2 | 2 | 152 | 3.56 | 3.45 | 3.67 | 0.71 |
| Responsibility | Current Personality | Group 3 | 1 | 176 | 3.60 | 3.49 | 3.71 | 0.74 |
| Responsibility | Current Personality | Group 3 | 2 | 159 | 3.62 | 3.51 | 3.73 | 0.72 |
| Responsibility | Ideal Personality | Group 1 | 1 | 174 | 4.26 | 4.17 | 4.35 | 0.58 |
| Responsibility | Ideal Personality | Group 1 | 2 | 155 | 4.29 | 4.19 | 4.38 | 0.60 |
| Neuroticism | Current Personality | Group 1 | 1 | 174 | 3.19 | 3.06 | 3.32 | 0.86 |
| Neuroticism | Current Personality | Group 1 | 2 | 155 | 3.13 | 2.99 | 3.26 | 0.85 |
| Neuroticism | Current Personality | Group 2 | 1 | 179 | 3.26 | 3.14 | 3.39 | 0.85 |
| Neuroticism | Current Personality | Group 2 | 2 | 152 | 3.10 | 2.97 | 3.24 | 0.85 |
| Neuroticism | Current Personality | Group 3 | 1 | 176 | 3.03 | 2.89 | 3.17 | 0.94 |
| Neuroticism | Current Personality | Group 3 | 2 | 159 | 3.04 | 2.89 | 3.19 | 0.96 |
| Neuroticism | Ideal Personality | Group 1 | 1 | 174 | 1.80 | 1.73 | 1.87 | 0.48 |
| Neuroticism | Ideal Personality | Group 1 | 2 | 155 | 1.82 | 1.75 | 1.90 | 0.48 |
| Anxiety | Current Personality | Group 1 | 1 | 174 | 3.53 | 3.39 | 3.67 | 0.94 |
| Anxiety | Current Personality | Group 1 | 2 | 155 | 3.50 | 3.36 | 3.65 | 0.91 |
| Anxiety | Current Personality | Group 2 | 1 | 179 | 3.60 | 3.47 | 3.73 | 0.90 |
| Anxiety | Current Personality | Group 2 | 2 | 152 | 3.50 | 3.34 | 3.66 | 0.98 |
| Anxiety | Current Personality | Group 3 | 1 | 176 | 3.41 | 3.26 | 3.56 | 1.00 |
| Anxiety | Current Personality | Group 3 | 2 | 159 | 3.42 | 3.27 | 3.58 | 1.00 |
| Anxiety | Ideal Personality | Group 1 | 1 | 174 | 1.90 | 1.81 | 1.99 | 0.62 |
| Anxiety | Ideal Personality | Group 1 | 2 | 155 | 1.97 | 1.87 | 2.07 | 0.62 |
| Depression | Current Personality | Group 1 | 1 | 174 | 3.09 | 2.94 | 3.24 | 1.00 |
| Depression | Current Personality | Group 1 | 2 | 155 | 3.01 | 2.86 | 3.17 | 1.00 |
| Depression | Current Personality | Group 2 | 1 | 179 | 3.18 | 3.03 | 3.32 | 0.98 |
| Depression | Current Personality | Group 2 | 2 | 152 | 2.94 | 2.79 | 3.10 | 0.99 |
| Depression | Current Personality | Group 3 | 1 | 176 | 2.83 | 2.67 | 2.98 | 1.05 |
| Depression | Current Personality | Group 3 | 2 | 159 | 2.86 | 2.69 | 3.03 | 1.09 |
| Depression | Ideal Personality | Group 1 | 1 | 174 | 1.51 | 1.43 | 1.59 | 0.54 |
| Depression | Ideal Personality | Group 1 | 2 | 155 | 1.54 | 1.45 | 1.62 | 0.55 |
| Volatility | Current Personality | Group 1 | 1 | 174 | 2.94 | 2.79 | 3.09 | 1.00 |
| Volatility | Current Personality | Group 1 | 2 | 155 | 2.87 | 2.71 | 3.02 | 1.00 |
| Volatility | Current Personality | Group 2 | 1 | 179 | 3.01 | 2.86 | 3.16 | 1.03 |
| Volatility | Current Personality | Group 2 | 2 | 152 | 2.87 | 2.72 | 3.02 | 0.95 |
| Volatility | Current Personality | Group 3 | 1 | 176 | 2.86 | 2.70 | 3.01 | 1.06 |
| Volatility | Current Personality | Group 3 | 2 | 159 | 2.83 | 2.67 | 3.00 | 1.07 |
| Volatility | Ideal Personality | Group 1 | 1 | 174 | 1.99 | 1.90 | 2.07 | 0.60 |
| Volatility | Ideal Personality | Group 1 | 2 | 155 | 1.96 | 1.87 | 2.05 | 0.58 |
| Openness | Current Personality | Group 1 | 1 | 174 | 3.74 | 3.64 | 3.83 | 0.61 |
| Openness | Current Personality | Group 1 | 2 | 155 | 3.77 | 3.68 | 3.86 | 0.60 |
| Openness | Current Personality | Group 2 | 1 | 179 | 3.70 | 3.61 | 3.80 | 0.67 |
| Openness | Current Personality | Group 2 | 2 | 152 | 3.77 | 3.66 | 3.87 | 0.66 |
| Openness | Current Personality | Group 3 | 1 | 176 | 3.83 | 3.74 | 3.93 | 0.64 |
| Openness | Current Personality | Group 3 | 2 | 159 | 3.84 | 3.74 | 3.94 | 0.63 |
| Openness | Ideal Personality | Group 1 | 1 | 174 | 4.12 | 4.04 | 4.20 | 0.55 |
| Openness | Ideal Personality | Group 1 | 2 | 155 | 4.14 | 4.05 | 4.23 | 0.54 |
| Curiosity | Current Personality | Group 1 | 1 | 174 | 3.88 | 3.77 | 3.98 | 0.69 |
| Curiosity | Current Personality | Group 1 | 2 | 155 | 3.93 | 3.82 | 4.04 | 0.70 |
| Curiosity | Current Personality | Group 2 | 1 | 179 | 3.88 | 3.77 | 3.99 | 0.74 |
| Curiosity | Current Personality | Group 2 | 2 | 152 | 3.97 | 3.84 | 4.09 | 0.76 |
| Curiosity | Current Personality | Group 3 | 1 | 176 | 4.01 | 3.90 | 4.11 | 0.72 |
| Curiosity | Current Personality | Group 3 | 2 | 159 | 4.01 | 3.91 | 4.12 | 0.68 |
| Curiosity | Ideal Personality | Group 1 | 1 | 174 | 4.15 | 4.05 | 4.25 | 0.68 |
| Curiosity | Ideal Personality | Group 1 | 2 | 155 | 4.13 | 4.03 | 4.23 | 0.64 |
| Aesthetic | Current Personality | Group 1 | 1 | 174 | 3.60 | 3.48 | 3.72 | 0.80 |
| Aesthetic | Current Personality | Group 1 | 2 | 155 | 3.64 | 3.51 | 3.76 | 0.82 |
| Aesthetic | Current Personality | Group 2 | 1 | 179 | 3.60 | 3.47 | 3.73 | 0.86 |
| Aesthetic | Current Personality | Group 2 | 2 | 152 | 3.63 | 3.50 | 3.76 | 0.83 |
| Aesthetic | Current Personality | Group 3 | 1 | 176 | 3.72 | 3.59 | 3.85 | 0.86 |
| Aesthetic | Current Personality | Group 3 | 2 | 159 | 3.72 | 3.59 | 3.85 | 0.85 |
| Aesthetic | Ideal Personality | Group 1 | 1 | 174 | 3.86 | 3.76 | 3.97 | 0.69 |
| Aesthetic | Ideal Personality | Group 1 | 2 | 155 | 3.95 | 3.84 | 4.06 | 0.70 |
| Imagination | Current Personality | Group 1 | 1 | 174 | 3.74 | 3.62 | 3.85 | 0.76 |
| Imagination | Current Personality | Group 1 | 2 | 155 | 3.75 | 3.63 | 3.87 | 0.75 |
| Imagination | Current Personality | Group 2 | 1 | 179 | 3.63 | 3.51 | 3.76 | 0.86 |
| Imagination | Current Personality | Group 2 | 2 | 152 | 3.71 | 3.58 | 3.85 | 0.84 |
| Imagination | Current Personality | Group 3 | 1 | 176 | 3.77 | 3.65 | 3.90 | 0.82 |
| Imagination | Current Personality | Group 3 | 2 | 159 | 3.79 | 3.67 | 3.92 | 0.80 |
| Imagination | Ideal Personality | Group 1 | 1 | 174 | 4.34 | 4.23 | 4.45 | 0.72 |
| Imagination | Ideal Personality | Group 1 | 2 | 155 | 4.34 | 4.24 | 4.45 | 0.66 |
Plotting mean-level changes on a descriptive level: Big Five dimensions and facets
ggplot(df_manip_check_forplot,
aes(y = m_trait, x = Time, shape = Group, color = Reference)) +
geom_point(position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(ymin = lwr95, ymax = upr95), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level (95% CI)") +
facet_wrap( ~ Trait, ncol = 4) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting mean-level changes on a descriptive level: Extraversion
ggplot(df_manip_check_forplot %>% filter(Trait=="Extraversion"),
aes(y = m_trait, x = Time, shape = Group, color = Reference)) +
geom_point(position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(ymin = lwr95, ymax = upr95), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level (95% CI)") +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting mean-level changes on a descriptive level: Agreeableness
ggplot(df_manip_check_forplot %>% filter(Trait=="Agreeableness"),
aes(y = m_trait, x = Time, shape = Group, color = Reference)) +
geom_point(position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(ymin = lwr95, ymax = upr95), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level (95% CI)") +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting mean-level changes on a descriptive level: Conscientiousness
ggplot(df_manip_check_forplot %>% filter(Trait=="Conscientiousness"),
aes(y = m_trait, x = Time, shape = Group, color = Reference)) +
geom_point(position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(ymin = lwr95, ymax = upr95), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level (95% CI)") +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting mean-level changes on a descriptive level: Neuroticism
ggplot(df_manip_check_forplot %>% filter(Trait=="Neuroticism"),
aes(y = m_trait, x = Time, shape = Group, color = Reference)) +
geom_point(position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(ymin = lwr95, ymax = upr95), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level (95% CI)") +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting mean-level changes on a descriptive level: Openness
ggplot(df_manip_check_forplot %>% filter(Trait=="Openness"),
aes(y = m_trait, x = Time, shape = Group, color = Reference)) +
geom_point(position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(ymin = lwr95, ymax = upr95), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level (95% CI)") +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Prepare data frame for plotting:
df_manip_check_wb_forplot <- df_sbsa3 %>%
mutate(time_d = time - 1) %>%
select(pid, time_d, group, swls, meaning, selfes, concept) %>%
pivot_longer(-c(pid, time_d, group),
cols_vary = "slowest",
names_to = c("trait"),
values_to = "score") %>%
mutate(trait = case_when(trait=="swls" ~ "Life Satisfaction", trait=="meaning" ~ "Meaning in Life",
trait=="selfes" ~ "Self-Esteem", trait=="concept" ~ "Self-Concept Clarity")) %>%
filter(!is.na(score)) %>%
group_by(trait, group, time_d) %>%
summarise(n = n(), m_trait = mean(score), sd_trait = sd(score)) %>%
ungroup() %>%
mutate(Group = factor(group, levels = c("Group 1", "Group 2", "Group 3"), labels = c("Group 1", "Group 2", "Group 3")),
Time = factor(time_d+1),
Trait = factor(trait,
levels = c("Life Satisfaction", "Meaning in Life", "Self-Esteem", "Self-Concept Clarity"), # group facets with each trait
labels = c("Life Satisfaction", "Meaning in Life", "Self-Esteem", "Self-Concept Clarity")),
lwr95 = m_trait - qnorm(0.975)*sd_trait/sqrt(n),
upr95 = m_trait + qnorm(0.975)*sd_trait/sqrt(n)) %>%
select(-c(time_d, group, trait))Table of all outcomes:
df_manip_check_wb_forplot %>% arrange(Trait) %>%
mutate_if(is.numeric, round, digits=2) %>%
select(Trait, Group, Time, n, m_trait, lwr95, upr95, sd_trait) %>%
kable(digits = 2)| Trait | Group | Time | n | m_trait | lwr95 | upr95 | sd_trait |
|---|---|---|---|---|---|---|---|
| Life Satisfaction | Group 1 | 1 | 174 | 2.75 | 2.60 | 2.90 | 1.01 |
| Life Satisfaction | Group 1 | 2 | 155 | 2.94 | 2.79 | 3.10 | 0.97 |
| Life Satisfaction | Group 2 | 1 | 179 | 2.62 | 2.48 | 2.77 | 0.99 |
| Life Satisfaction | Group 2 | 2 | 151 | 2.87 | 2.71 | 3.03 | 1.00 |
| Life Satisfaction | Group 3 | 1 | 176 | 2.94 | 2.79 | 3.09 | 1.00 |
| Life Satisfaction | Group 3 | 2 | 158 | 2.95 | 2.79 | 3.11 | 1.02 |
| Meaning in Life | Group 1 | 1 | 174 | 3.76 | 3.59 | 3.93 | 1.14 |
| Meaning in Life | Group 1 | 2 | 155 | 3.91 | 3.73 | 4.09 | 1.14 |
| Meaning in Life | Group 2 | 1 | 179 | 3.59 | 3.43 | 3.76 | 1.11 |
| Meaning in Life | Group 2 | 2 | 151 | 3.73 | 3.56 | 3.91 | 1.10 |
| Meaning in Life | Group 3 | 1 | 176 | 3.77 | 3.59 | 3.94 | 1.17 |
| Meaning in Life | Group 3 | 2 | 158 | 3.90 | 3.71 | 4.09 | 1.22 |
| Self-Esteem | Group 1 | 1 | 174 | 3.33 | 3.20 | 3.46 | 0.88 |
| Self-Esteem | Group 1 | 2 | 155 | 3.44 | 3.30 | 3.57 | 0.85 |
| Self-Esteem | Group 2 | 1 | 179 | 3.14 | 3.02 | 3.26 | 0.83 |
| Self-Esteem | Group 2 | 2 | 151 | 3.33 | 3.19 | 3.47 | 0.88 |
| Self-Esteem | Group 3 | 1 | 176 | 3.44 | 3.30 | 3.58 | 0.93 |
| Self-Esteem | Group 3 | 2 | 158 | 3.45 | 3.30 | 3.61 | 0.99 |
| Self-Concept Clarity | Group 1 | 1 | 174 | 3.10 | 2.97 | 3.23 | 0.88 |
| Self-Concept Clarity | Group 1 | 2 | 155 | 3.29 | 3.15 | 3.42 | 0.88 |
| Self-Concept Clarity | Group 2 | 1 | 179 | 3.03 | 2.91 | 3.15 | 0.83 |
| Self-Concept Clarity | Group 2 | 2 | 151 | 3.13 | 3.00 | 3.26 | 0.83 |
| Self-Concept Clarity | Group 3 | 1 | 176 | 3.16 | 3.03 | 3.30 | 0.91 |
| Self-Concept Clarity | Group 3 | 2 | 158 | 3.24 | 3.10 | 3.38 | 0.91 |
Plotting mean-level changes on a descriptive level: all four well-being aspects
ggplot(df_manip_check_wb_forplot,
aes(y = m_trait, x = Time, shape = Group)) +
geom_point(position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(ymin = lwr95, ymax = upr95), width=.3, position=position_dodge(0.3)) +
ylab("Mean Well-Being Level (95% CI)") +
facet_wrap( ~ Trait, ncol = 2) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
For Study 3, most of these moedls are only possible in Group 1 which answered both current and ideal personality trait items.
I only left the heading “confirmatory” here for better comparability with Study 1 and Study 2 analyses. The preregistered, confirmatory analyses for Study 3 are found below in section 7!
All four psychological well-being indicators will be positively correlated with a greater similarity between current- and ideal personality-ratings of personality.
To examine this at the level of overall profiles, we will compute the correlations between the psychological well-being indicators and the Fisher z transformed correlations between the facet- and item-level real-ideal personality-profiles. To examine this at the level of individual traits, we will compute the correlation between psychological well-being indicators and the squared difference between current- and ideal personality rating for each Big Five trait and facet.
In Study 3, we can only examine this in Group 3 because the other groups only received current personality items.
Computations:
cormat_profile <- cor(df_sbsa3[, c("swls", "meaning", "selfes", "concept",
"profile_corr_item_z", "profile_corr_facet_z")], use = "pairwise.complete.obs")
rownames(cormat_profile) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
"Item-Level Profile Corr (z)", "Facet-Level Profile Corr (z)")
colnames(cormat_profile) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
"Item-Level Profile Corr (z)", "Facet-Level Profile Corr (z)")corrplot(cormat_profile, type = "lower", order = "original", tl.col = "black", tl.srt = 10,
addCoef.col ='black', number.cex = 0.7, diag = FALSE) # also add numbers
As a table with confidence intervals - profile similarity:
bind_rows(
bind_cols(
cormat_profile[5, c(1:4)] %>% as_tibble() %>%
mutate(wellbeing = c("swls", "meaning", "selfes", "concept"),
profile_sim = rep("item-level", 4)) %>%
select(profile_sim, wellbeing, corr = value),
correlation::cor_to_ci(cormat_profile[5, c(1:4)], n = (df_sbsa3 %>% filter(!is.na(swls))
%>% tally() %>% pull())) %>% as_tibble(),
correlation::cor_to_p(cormat_profile[5, c(1:4)], n = (df_sbsa3 %>% filter(!is.na(swls))
%>% tally() %>% pull())) %>% as_tibble() %>% select(p)
),
bind_cols(
cormat_profile[6, c(1:4)] %>% as_tibble() %>%
mutate(wellbeing = c("swls", "meaning", "selfes", "concept"),
profile_sim = rep("facet-level", 4)) %>%
select(profile_sim, wellbeing, corr = value),
correlation::cor_to_ci(cormat_profile[6, c(1:4)], n = (df_sbsa3 %>% filter(!is.na(swls))
%>% tally() %>% pull())) %>% as_tibble(),
correlation::cor_to_p(cormat_profile[6, c(1:4)], n = (df_sbsa3 %>% filter(!is.na(swls))
%>% tally() %>% pull())) %>% as_tibble() %>% select(p)
)
) %>% kable(digits = 3)| profile_sim | wellbeing | corr | CI_low | CI_high | p |
|---|---|---|---|---|---|
| item-level | swls | 0.435 | 0.383 | 0.484 | 0 |
| item-level | meaning | 0.401 | 0.348 | 0.452 | 0 |
| item-level | selfes | 0.602 | 0.561 | 0.641 | 0 |
| item-level | concept | 0.581 | 0.538 | 0.620 | 0 |
| facet-level | swls | 0.449 | 0.398 | 0.498 | 0 |
| facet-level | meaning | 0.431 | 0.379 | 0.480 | 0 |
| facet-level | selfes | 0.640 | 0.602 | 0.676 | 0 |
| facet-level | concept | 0.614 | 0.573 | 0.651 | 0 |
Computations:
# T1
cormat_profile_gr1_t1 <- cor((df_sbsa3 %>% filter(group=="Group 1" & time==1))[, c("swls", "meaning", "selfes", "concept", "profile_corr_item_z", "profile_corr_facet_z")], use = "pairwise.complete.obs")
rownames(cormat_profile_gr1_t1) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
"Item-Level Profile Corr (z)", "Facet-Level Profile Corr (z)")
colnames(cormat_profile_gr1_t1) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
"Item-Level Profile Corr (z)", "Facet-Level Profile Corr (z)")
# T2
cormat_profile_gr1_t2 <- cor((df_sbsa3 %>% filter(group=="Group 1" & time==2))[, c("swls", "meaning", "selfes", "concept", "profile_corr_item_z", "profile_corr_facet_z")], use = "pairwise.complete.obs")
rownames(cormat_profile_gr1_t2) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
"Item-Level Profile Corr (z)", "Facet-Level Profile Corr (z)")
colnames(cormat_profile_gr1_t2) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
"Item-Level Profile Corr (z)", "Facet-Level Profile Corr (z)")T1
First assessment
corrplot(cormat_profile_gr1_t1, type = "lower", order = "original", tl.col = "black", tl.srt = 10,
addCoef.col ='black', number.cex = 0.7, diag = FALSE) # also add numbers
T2
Second assessment
corrplot(cormat_profile_gr1_t2, type = "lower", order = "original", tl.col = "black", tl.srt = 10,
addCoef.col ='black', number.cex = 0.7, diag = FALSE) # also add numbers
Computations:
cormat_sqtraits <- cor(df_sbsa3[, c("swls", "meaning", "selfes", "concept",
paste0(str_trunc(names(b5_vars)[1:5], 5, ellipsis = ""), "_sqdiff"))],
use = "pairwise.complete.obs")
rownames(cormat_sqtraits) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
str_to_title(names(b5_vars)[1:5]))
colnames(cormat_sqtraits) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
str_to_title(names(b5_vars)[1:5]))
cormat_sqfacets <- cor(df_sbsa3[, c("swls", "meaning", "selfes", "concept",
paste0(str_trunc(names(b5_vars)[6:20], 5, ellipsis = ""), "_sqdiff"))],
use = "pairwise.complete.obs")
rownames(cormat_sqfacets) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
str_to_title(names(b5_vars)[6:20]))
colnames(cormat_sqfacets) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
str_to_title(names(b5_vars)[6:20]))Big Five traits
corrplot(cormat_sqtraits, type = "lower", order = "original", tl.col = "black", tl.srt = 10,
addCoef.col ='black', number.cex = 0.7, diag = FALSE) # also add numbers
As a table with confidence intervals - squared differences:
cormat_sqtraits[c(5:9), c(1:4)] %>% as_tibble() %>%
mutate(personality = names(b5_vars)[1:5]) %>%
rename(swls = `Life Satisfaction`, meaning = `Meaning in Life`,
selfes = `Self-Esteem`, concept = `Self Concept Clarity`) %>%
pivot_longer(1:4, names_to = "wellbeing", values_to = "corr") %>%
mutate(CI_low = correlation::cor_to_ci(corr, n = (df_sbsa3 %>% filter(!is.na(swls))
%>% tally() %>% pull())) %>%
as_tibble() %>% pull(CI_low),
CI_high = correlation::cor_to_ci(corr, n = (df_sbsa3 %>% filter(!is.na(swls))
%>% tally() %>% pull())) %>%
as_tibble() %>% pull(CI_high),
p = correlation::cor_to_p(corr, n = (df_sbsa3 %>% filter(!is.na(swls))
%>% tally() %>% pull())) %>%
as_tibble() %>% pull(p)) %>%
kable(digits = 3)| personality | wellbeing | corr | CI_low | CI_high | p |
|---|---|---|---|---|---|
| extraversion | swls | -0.256 | -0.314 | -0.197 | 0.000 |
| extraversion | meaning | -0.328 | -0.382 | -0.271 | 0.000 |
| extraversion | selfes | -0.388 | -0.440 | -0.334 | 0.000 |
| extraversion | concept | -0.310 | -0.365 | -0.252 | 0.000 |
| agreeableness | swls | -0.213 | -0.272 | -0.153 | 0.000 |
| agreeableness | meaning | -0.183 | -0.243 | -0.122 | 0.000 |
| agreeableness | selfes | -0.241 | -0.299 | -0.181 | 0.000 |
| agreeableness | concept | -0.289 | -0.345 | -0.231 | 0.000 |
| conscientiousness | swls | -0.296 | -0.351 | -0.238 | 0.000 |
| conscientiousness | meaning | -0.060 | -0.122 | 0.002 | 0.059 |
| conscientiousness | selfes | -0.291 | -0.347 | -0.233 | 0.000 |
| conscientiousness | concept | -0.217 | -0.275 | -0.157 | 0.000 |
| neuroticism | swls | -0.399 | -0.450 | -0.346 | 0.000 |
| neuroticism | meaning | -0.386 | -0.437 | -0.331 | 0.000 |
| neuroticism | selfes | -0.497 | -0.543 | -0.449 | 0.000 |
| neuroticism | concept | -0.442 | -0.490 | -0.390 | 0.000 |
| openness | swls | -0.083 | -0.144 | -0.020 | 0.009 |
| openness | meaning | -0.068 | -0.130 | -0.006 | 0.032 |
| openness | selfes | -0.115 | -0.176 | -0.053 | 0.000 |
| openness | concept | -0.016 | -0.079 | 0.046 | 0.606 |
Big Five facets
corrplot(cormat_sqfacets, type = "lower", order = "original", tl.col = "black", tl.srt = 10,
addCoef.col ='black', number.cex = 0.6, diag = FALSE) # also add numbers
Only looking at the trait level for now!
Computations:
# T1
cormat_sqtraits_gr1_t1 <- cor((df_sbsa3 %>% filter(group=="Group 1" & time==1))[, c("swls", "meaning", "selfes", "concept",
paste0(str_trunc(names(b5_vars)[1:5], 5, ellipsis = ""), "_sqdiff"))],
use = "pairwise.complete.obs")
rownames(cormat_sqtraits_gr1_t1) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
str_to_title(names(b5_vars)[1:5]))
colnames(cormat_sqtraits_gr1_t1) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
str_to_title(names(b5_vars)[1:5]))
# T2
cormat_sqtraits_gr1_t2 <- cor((df_sbsa3 %>% filter(group=="Group 1" & time==2))[, c("swls", "meaning", "selfes", "concept",
paste0(str_trunc(names(b5_vars)[1:5], 5, ellipsis = ""), "_sqdiff"))],
use = "pairwise.complete.obs")
rownames(cormat_sqtraits_gr1_t2) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
str_to_title(names(b5_vars)[1:5]))
colnames(cormat_sqtraits_gr1_t2) <- c("Life Satisfaction", "Meaning in Life", "Self-Esteem" , "Self Concept Clarity",
str_to_title(names(b5_vars)[1:5]))T1
First assessment:
corrplot(cormat_sqtraits_gr1_t1, type = "lower", order = "original", tl.col = "black", tl.srt = 10,
addCoef.col ='black', number.cex = 0.7, diag = FALSE) # also add numbers
T2
Second assessment:
corrplot(cormat_sqtraits_gr1_t2, type = "lower", order = "original", tl.col = "black", tl.srt = 10,
addCoef.col ='black', number.cex = 0.7, diag = FALSE) # also add numbers
Additional, exploratory analysis: Are the relationships of the squared trait differences with well-being similar when we control for initial trait levels (at T1)?
Example with extraversion:
cor(df_sbsa3 %>% filter(group=="Group 1" & time==1) %>% pull(swls),
df_sbsa3 %>% filter(group=="Group 1" & time==1) %>% pull(extra_sqdiff),
use = "pairwise.complete.obs") %>% round(2)[1] -0.27# basic model to replicate the correlation
mod_ex_1 <- lm(swls ~ extra_sqdiff, data = df_sbsa3 %>% filter(group=="Group 1" & time==1))
summary(mod_ex_1)$coef Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.989693 0.09815233 30.459725 3.331613e-71
extra_sqdiff -0.135079 0.03689506 -3.661167 3.338466e-04effectsize::standardize_parameters(mod_ex_1)# Standardization method: refit
Parameter | Std. Coef. | 95% CI
------------------------------------------
(Intercept) | 3.23e-16 | [-0.14, 0.14]
extra_sqdiff | -0.27 | [-0.41, -0.12]# adjusted model controlling for initial trait level
mod_ex_2 <- lm(swls ~ extra_sqdiff + extra_comb_curr,
data = df_sbsa3 %>% filter(group=="Group 1" & time==1))
summary(mod_ex_2)$coef Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.47215535 0.43334305 3.397205 0.0008467750
extra_sqdiff -0.01786883 0.04837127 -0.369410 0.7122791641
extra_comb_curr 0.46323156 0.12906599 3.589106 0.0004331415effectsize::standardize_parameters(mod_ex_2)# Standardization method: refit
Parameter | Std. Coef. | 95% CI
--------------------------------------------
(Intercept) | 3.42e-16 | [-0.14, 0.14]
extra_sqdiff | -0.04 | [-0.23, 0.15]
extra_comb_curr | 0.35 | [ 0.16, 0.54]Only T1:
# prepare df
df_check_adjusted <- df_sbsa3 %>%
# reshape with well-being outcomes and current trait levels
filter(group=="Group 1" & time==1) %>%
select(pid, group,
swls, meaning, selfes, concept,
paste0(str_trunc(names(b5_vars), 5, ellipsis = ""), "_comb_curr")) %>%
pivot_longer(ends_with("_comb_curr"),
names_to = "test", names_prefix = "facet", values_to = "initial", values_drop_na = TRUE) %>%
mutate(test = sub("_.*", "", test)) %>%
left_join(
df_sbsa3 %>%
# reshape with squared differences -> add to previous
filter(group=="Group 1" & time==1) %>%
select(pid, group, ends_with("_sqdiff")) %>%
pivot_longer(ends_with("_sqdiff"),
names_to = "test", names_prefix = "facet", values_to = "score", values_drop_na = TRUE) %>%
mutate(test = sub("_.*", "", test))
)
# run models
# swls
df_check_adjusted_swls <- df_check_adjusted %>%
group_nest(test) %>%
mutate(lm_mods = map(data, ~lm(score ~ swls + initial, data = .x))) %>%
pull(lm_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_check_adjusted_swls_unlist <- as.data.frame(summary(df_check_adjusted_swls[[1]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_swls[[1]])$Std_Coefficient) # standardized beta
for (i in 2:length(df_check_adjusted_swls)) {
df_check_adjusted_swls_unlist <-
bind_rows(df_check_adjusted_swls_unlist,
as.data.frame(summary(df_check_adjusted_swls[[i]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_swls[[i]])$Std_Coefficient))
}
df_check_adjusted_swls_unlist <- df_check_adjusted_swls_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=3),
term = c(rep(c("Intercept", "swls", "initial"), 20))) %>%
rename(p = `Pr(>|t|)`, std_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, d, std_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars))))
# meaning
df_check_adjusted_meaning <- df_check_adjusted %>%
group_nest(test) %>%
mutate(lm_mods = map(data, ~lm(score ~ meaning + initial, data = .x))) %>%
pull(lm_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_check_adjusted_meaning_unlist <- as.data.frame(summary(df_check_adjusted_meaning[[1]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_meaning[[1]])$Std_Coefficient) # standardized beta
for (i in 2:length(df_check_adjusted_meaning)) {
df_check_adjusted_meaning_unlist <-
bind_rows(df_check_adjusted_meaning_unlist,
as.data.frame(summary(df_check_adjusted_meaning[[i]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_meaning[[i]])$Std_Coefficient))
}
df_check_adjusted_meaning_unlist <- df_check_adjusted_meaning_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=3),
term = c(rep(c("Intercept", "meaning", "initial"), 20))) %>%
rename(p = `Pr(>|t|)`, std_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, d, std_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars))))
# selfes
df_check_adjusted_selfes <- df_check_adjusted %>%
group_nest(test) %>%
mutate(lm_mods = map(data, ~lm(score ~ selfes + initial, data = .x))) %>%
pull(lm_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_check_adjusted_selfes_unlist <- as.data.frame(summary(df_check_adjusted_selfes[[1]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_selfes[[1]])$Std_Coefficient) # standardized beta
for (i in 2:length(df_check_adjusted_selfes)) {
df_check_adjusted_selfes_unlist <-
bind_rows(df_check_adjusted_selfes_unlist,
as.data.frame(summary(df_check_adjusted_selfes[[i]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_selfes[[i]])$Std_Coefficient))
}
df_check_adjusted_selfes_unlist <- df_check_adjusted_selfes_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=3),
term = c(rep(c("Intercept", "selfes", "initial"), 20))) %>%
rename(p = `Pr(>|t|)`, std_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, d, std_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars))))
# concept
df_check_adjusted_concept <- df_check_adjusted %>%
group_nest(test) %>%
mutate(lm_mods = map(data, ~lm(score ~ concept + initial, data = .x))) %>%
pull(lm_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_check_adjusted_concept_unlist <- as.data.frame(summary(df_check_adjusted_concept[[1]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_concept[[1]])$Std_Coefficient) # standardized beta
for (i in 2:length(df_check_adjusted_concept)) {
df_check_adjusted_concept_unlist <-
bind_rows(df_check_adjusted_concept_unlist,
as.data.frame(summary(df_check_adjusted_concept[[i]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_concept[[i]])$Std_Coefficient))
}
df_check_adjusted_concept_unlist <- df_check_adjusted_concept_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=3),
term = c(rep(c("Intercept", "concept", "initial"), 20))) %>%
rename(p = `Pr(>|t|)`, std_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, d, std_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars))))Does the relationship between the squared difference in one trait (current vs. ideal) and well-being aspects still hold if we control for the initial trait level?
Combine as one table:
kable(df_check_adjusted_swls_unlist %>%
filter(!term %in% c("Intercept", "initial")) %>%
select(outcome, d_swls = d, p_ls = p, sig_ls = sig) %>%
left_join(
df_check_adjusted_meaning_unlist %>%
filter(!term %in% c("Intercept", "initial")) %>%
select(outcome, d_meaning = d, p_m = p, sig_m = sig)
) %>%
left_join(
df_check_adjusted_selfes_unlist %>%
filter(!term %in% c("Intercept", "initial")) %>%
select(outcome, d_selfes = d, p_se = p, sig_se = sig)
) %>%
left_join(
df_check_adjusted_concept_unlist %>%
filter(!term %in% c("Intercept", "initial")) %>%
select(outcome, d_concept = d, p_c = p, sig_c = sig)
),
digits = 3)| outcome | d_swls | p_ls | sig_ls | d_meaning | p_m | sig_m | d_selfes | p_se | sig_se | d_concept | p_c | sig_c |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| extraversion | -0.022 | 0.712 | -0.108 | 0.074 | -0.117 | 0.063 | -0.081 | 0.178 | ||||
| agreeableness | -0.110 | 0.114 | -0.109 | 0.110 | -0.157 | 0.025 | * | -0.149 | 0.035 | * | ||
| conscientiousness | -0.088 | 0.069 | 0.003 | 0.956 | 0.016 | 0.759 | 0.099 | 0.048 | * | |||
| neuroticism | -0.039 | 0.429 | 0.022 | 0.667 | 0.046 | 0.412 | 0.135 | 0.016 | * | |||
| openness | -0.085 | 0.205 | -0.019 | 0.781 | -0.079 | 0.245 | 0.039 | 0.567 | ||||
| sociability | -0.017 | 0.794 | -0.114 | 0.065 | -0.141 | 0.024 | * | -0.118 | 0.053 | |||
| assertiveness | 0.024 | 0.702 | -0.070 | 0.280 | -0.057 | 0.390 | -0.059 | 0.357 | ||||
| energy | -0.055 | 0.321 | -0.072 | 0.196 | -0.047 | 0.429 | -0.042 | 0.455 | ||||
| compassion | -0.183 | 0.015 | * | -0.140 | 0.062 | -0.192 | 0.010 | * | -0.101 | 0.182 | ||
| respectfulness | -0.068 | 0.264 | -0.067 | 0.264 | -0.054 | 0.380 | -0.071 | 0.255 | ||||
| trust | -0.059 | 0.384 | -0.055 | 0.412 | -0.099 | 0.142 | -0.113 | 0.096 | ||||
| organization | -0.116 | 0.015 | * | -0.075 | 0.111 | -0.034 | 0.487 | -0.034 | 0.485 | |||
| productiveness | -0.105 | 0.020 | * | 0.017 | 0.707 | -0.010 | 0.831 | 0.089 | 0.056 | |||
| responsibility | -0.066 | 0.258 | -0.008 | 0.887 | -0.044 | 0.463 | 0.099 | 0.095 | ||||
| anxiety | -0.016 | 0.763 | 0.072 | 0.190 | 0.008 | 0.894 | 0.097 | 0.095 | ||||
| depression | -0.075 | 0.134 | -0.087 | 0.102 | 0.008 | 0.896 | 0.088 | 0.118 | ||||
| volatility | -0.035 | 0.507 | 0.041 | 0.434 | 0.057 | 0.307 | 0.118 | 0.038 | * | |||
| curiosity | -0.057 | 0.418 | 0.052 | 0.460 | -0.067 | 0.345 | -0.051 | 0.469 | ||||
| aesthetic | -0.051 | 0.463 | -0.025 | 0.720 | -0.093 | 0.180 | -0.008 | 0.903 | ||||
| imagination | -0.096 | 0.137 | -0.072 | 0.273 | -0.028 | 0.672 | 0.006 | 0.933 |
Across all time points, using a mixed model:
# prepare df
df_check_adjusted_all <- df_sbsa3 %>%
# reshape with well-being outcomes and current trait levels
filter(group=="Group 1") %>%
select(pid, group, time,
swls, meaning, selfes, concept,
paste0(str_trunc(names(b5_vars), 5, ellipsis = ""), "_sqdiff")) %>%
pivot_longer(ends_with("_sqdiff"),
names_to = "test", names_prefix = "facet", values_to = "score", values_drop_na = TRUE) %>%
mutate(test = sub("_.*", "", test)) %>%
left_join(
df_sbsa3 %>%
# reshape with squared differences -> add to previous
filter(group=="Group 1" & time==1) %>%
select(pid, group, ends_with("_comb_curr")) %>%
pivot_longer(ends_with("_comb_curr"),
names_to = "test", names_prefix = "facet", values_to = "initial", values_drop_na = TRUE) %>%
mutate(test = sub("_.*", "", test))
)
# run models
# swls
df_check_adjusted_swls_all <- df_check_adjusted_all %>%
group_nest(test) %>%
mutate(lm_mods = map(data, ~lmerTest::lmer(score ~ swls + initial + (1 | pid), data = .x))) %>% # groupm intercept model
pull(lm_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_check_adjusted_swls_all_unlist <- as.data.frame(summary(df_check_adjusted_swls_all[[1]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_swls_all[[1]])$Std_Coefficient) # standardized beta
for (i in 2:length(df_check_adjusted_swls_all)) {
df_check_adjusted_swls_all_unlist <-
bind_rows(df_check_adjusted_swls_all_unlist,
as.data.frame(summary(df_check_adjusted_swls_all[[i]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_swls_all[[i]])$Std_Coefficient))
}
df_check_adjusted_swls_all_unlist <- df_check_adjusted_swls_all_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=3),
term = c(rep(c("Intercept", "swls", "initial"), 20))) %>%
rename(p = `Pr(>|t|)`, std_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, d, std_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars))))
# meaning
df_check_adjusted_meaning_all <- df_check_adjusted_all %>%
group_nest(test) %>%
mutate(lm_mods = map(data, ~lmerTest::lmer(score ~ meaning + initial + (1 | pid), data = .x))) %>% # groupm intercept model
pull(lm_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_check_adjusted_meaning_all_unlist <- as.data.frame(summary(df_check_adjusted_meaning_all[[1]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_meaning_all[[1]])$Std_Coefficient)
for (i in 2:length(df_check_adjusted_meaning_all)) {
df_check_adjusted_meaning_all_unlist <-
bind_rows(df_check_adjusted_meaning_all_unlist,
as.data.frame(summary(df_check_adjusted_meaning_all[[i]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_meaning_all[[i]])$Std_Coefficient))
}
df_check_adjusted_meaning_all_unlist <- df_check_adjusted_meaning_all_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=3),
term = c(rep(c("Intercept", "meaning", "initial"), 20))) %>%
rename(p = `Pr(>|t|)`, std_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, d, std_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars))))
# selfes
df_check_adjusted_selfes_all <- df_check_adjusted_all %>%
group_nest(test) %>%
mutate(lm_mods = map(data, ~lmerTest::lmer(score ~ selfes + initial + (1 | pid), data = .x))) %>% # groupm intercept model
pull(lm_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_check_adjusted_selfes_all_unlist <- as.data.frame(summary(df_check_adjusted_selfes_all[[1]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_selfes_all[[1]])$Std_Coefficient)
for (i in 2:length(df_check_adjusted_selfes_all)) {
df_check_adjusted_selfes_all_unlist <-
bind_rows(df_check_adjusted_selfes_all_unlist,
as.data.frame(summary(df_check_adjusted_selfes_all[[i]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_selfes_all[[i]])$Std_Coefficient))
}
df_check_adjusted_selfes_all_unlist <- df_check_adjusted_selfes_all_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=3),
term = c(rep(c("Intercept", "selfes", "initial"), 20))) %>%
rename(p = `Pr(>|t|)`, std_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, d, std_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars))))
# concept
df_check_adjusted_concept_all <- df_check_adjusted_all %>%
group_nest(test) %>%
mutate(lm_mods = map(data, ~lmerTest::lmer(score ~ concept + initial + (1 | pid), data = .x))) %>% # groupm intercept model
pull(lm_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_check_adjusted_concept_all_unlist <- as.data.frame(summary(df_check_adjusted_concept_all[[1]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_concept_all[[1]])$Std_Coefficient)
for (i in 2:length(df_check_adjusted_concept_all)) {
df_check_adjusted_concept_all_unlist <-
bind_rows(df_check_adjusted_concept_all_unlist,
as.data.frame(summary(df_check_adjusted_concept_all[[i]])$coefficients) %>% as_tibble() %>%
mutate(d = effectsize::standardize_parameters(df_check_adjusted_concept_all[[i]])$Std_Coefficient))
}
df_check_adjusted_concept_all_unlist <- df_check_adjusted_concept_all_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=3),
term = c(rep(c("Intercept", "concept", "initial"), 20))) %>%
rename(p = `Pr(>|t|)`, std_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, d, std_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars))))Does the relationship between the squared difference in one trait (current vs. ideal) and well-being aspects still hold if we control for the initial trait level?
Combine as one table:
kable(df_check_adjusted_swls_all_unlist %>%
filter(!term %in% c("Intercept", "initial")) %>%
select(outcome, d_swls = d, p_ls = p, sig_ls = sig) %>%
left_join(
df_check_adjusted_meaning_all_unlist %>%
filter(!term %in% c("Intercept", "initial")) %>%
select(outcome, d_meaning = d, p_m = p, sig_m = sig)
) %>%
left_join(
df_check_adjusted_selfes_all_unlist %>%
filter(!term %in% c("Intercept", "initial")) %>%
select(outcome, d_selfes = d, p_se = p, sig_se = sig)
) %>%
left_join(
df_check_adjusted_concept_all_unlist %>%
filter(!term %in% c("Intercept", "initial")) %>%
select(outcome, d_concept = d, p_c = p, sig_c = sig)
),
digits = 3)| outcome | d_swls | p_ls | sig_ls | d_meaning | p_m | sig_m | d_selfes | p_se | sig_se | d_concept | p_c | sig_c |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| extraversion | -0.028 | 0.566 | -0.118 | 0.016 | * | -0.098 | 0.057 | -0.112 | 0.018 | * | ||
| agreeableness | -0.095 | 0.092 | -0.115 | 0.041 | * | -0.106 | 0.064 | -0.188 | 0.001 | *** | ||
| conscientiousness | -0.068 | 0.098 | 0.001 | 0.973 | -0.006 | 0.890 | 0.013 | 0.749 | ||||
| neuroticism | -0.080 | 0.058 | -0.037 | 0.386 | -0.066 | 0.168 | -0.017 | 0.699 | ||||
| openness | -0.058 | 0.284 | -0.021 | 0.704 | -0.067 | 0.224 | 0.030 | 0.576 | ||||
| sociability | -0.056 | 0.288 | -0.125 | 0.016 | * | -0.137 | 0.010 | ** | -0.148 | 0.003 | ** | |
| assertiveness | 0.029 | 0.586 | -0.072 | 0.178 | -0.066 | 0.226 | -0.092 | 0.076 | ||||
| energy | -0.048 | 0.316 | -0.084 | 0.078 | -0.046 | 0.364 | -0.066 | 0.159 | ||||
| compassion | -0.135 | 0.026 | * | -0.114 | 0.062 | -0.130 | 0.033 | * | -0.155 | 0.010 | ** | |
| respectfulness | -0.074 | 0.148 | -0.107 | 0.036 | * | -0.076 | 0.149 | -0.161 | 0.002 | ** | ||
| trust | -0.114 | 0.040 | * | -0.070 | 0.202 | -0.120 | 0.032 | * | -0.125 | 0.023 | * | |
| organization | -0.082 | 0.049 | * | -0.065 | 0.120 | -0.031 | 0.465 | -0.053 | 0.202 | |||
| productiveness | -0.143 | 0.001 | *** | -0.027 | 0.511 | -0.079 | 0.078 | -0.030 | 0.492 | |||
| responsibility | -0.032 | 0.496 | 0.049 | 0.286 | -0.036 | 0.449 | 0.027 | 0.548 | ||||
| anxiety | -0.059 | 0.200 | -0.006 | 0.893 | -0.091 | 0.065 | -0.041 | 0.393 | ||||
| depression | -0.162 | 0.000 | *** | -0.168 | 0.000 | *** | -0.129 | 0.015 | * | -0.027 | 0.558 | |
| volatility | -0.045 | 0.330 | 0.003 | 0.939 | -0.012 | 0.806 | 0.014 | 0.780 | ||||
| curiosity | -0.071 | 0.208 | 0.014 | 0.808 | -0.095 | 0.096 | -0.069 | 0.217 | ||||
| aesthetic | -0.009 | 0.870 | -0.009 | 0.870 | -0.052 | 0.350 | 0.005 | 0.923 | ||||
| imagination | -0.090 | 0.089 | -0.110 | 0.045 | * | -0.049 | 0.369 | -0.025 | 0.639 |
Both groups will increase in all four psychological well-being indicators.
We will test the mean-level difference between baseline and follow up using a latent change model.
Fit model:
# Code snippets adapted from Kievit et al. (2018) -- CC-BY -- https://doi.org/10.1016/j.dcn.2017.11.007
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_swls_hyp2 <- '
swls_t1 =~ 1*sw06_01_t1 + lamb2*sw06_02_t1 + lamb3*sw06_03_t1 + lamb4*sw06_04_t1 # This specifies the measurement model for swls_t1
swls_t2 =~ 1*sw06_01_t2 + lamb2*sw06_02_t2 + lamb3*sw06_03_t2 + lamb4*sw06_04_t2 # This specifies the measurement model for swls_t2 with the equality constrained factor loadings
swls_t2 ~ 1*swls_t1 # This parameter regresses swls_t2 perfectly on swls_t1
d_swls_1 =~ 1*swls_t2 # This defines the latent change score factor as measured perfectly by scores on swls_t2
swls_t2 ~ 0*1 # This line constrains the intercept of swls_t2 to 0
swls_t2 ~~ 0*swls_t2 # This fixes the variance of swls_t2 to 0
d_swls_1 ~ 1 # This estimates the intercept of the change score
swls_t1 ~ 1 # This estimates the intercept of swls_t1
d_swls_1 ~~ d_swls_1 # This estimates the variance of the change scores
swls_t1 ~~ swls_t1 # This estimates the variance of the swls_t1
d_swls_1 ~~ swls_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
sw06_01_t1 ~~ sw06_01_t2 # This allows residual covariance on indicator X1 across T1 and T2
sw06_02_t1 ~~ sw06_02_t2 # This allows residual covariance on indicator X2 across T1 and T2
sw06_03_t1 ~~ sw06_03_t2 # This allows residual covariance on indicator X3 across T1 and T2
sw06_04_t1 ~~ sw06_04_t2 # This allows residual covariance on indicator X4 across T1 and T2
sw06_01_t1 ~~ res1*sw06_01_t1 # This allows residual variance on indicator X1 at T1
sw06_02_t1 ~~ res2*sw06_02_t1 # This allows residual variance on indicator X2 at T1
sw06_03_t1 ~~ res3*sw06_03_t1 # This allows residual variance on indicator X3 at T1
sw06_04_t1 ~~ res4*sw06_04_t1 # This allows residual variance on indicator X4 at T1
sw06_01_t2 ~~ res1*sw06_01_t2 # This allows residual variance on indicator X1 at T2
sw06_02_t2 ~~ res2*sw06_02_t2 # This allows residual variance on indicator X2 at T2
sw06_03_t2 ~~ res3*sw06_03_t2 # This allows residual variance on indicator X3 at T2
sw06_04_t2 ~~ res4*sw06_04_t2 # This allows residual variance on indicator X4 at T2
sw06_01_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
sw06_02_t1 ~ m2*1 # This estimates the intercept of X2 at T1
sw06_03_t1 ~ m3*1 # This estimates the intercept of X3 at T1
sw06_04_t1 ~ m4*1 # This estimates the intercept of X4 at T1
sw06_01_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
sw06_02_t2 ~ m2*1 # This estimates the intercept of X2 at T2
sw06_03_t2 ~ m3*1 # This estimates the intercept of X3 at T2
sw06_04_t2 ~ m4*1 # This estimates the intercept of X4 at T2
'
fit_mi_lcs_swls_hyp2 <- lavaan(mi_lcs_swls_hyp2, data=df_sbsa3_wide_wb, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_swls_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_swls_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 19 | 10229.01 | 10310.2 | 0.989 | 0.047 | 0.028 |
# parameters of interest
params_lcs_swls_hyp2 <- broom::tidy(fit_mi_lcs_swls_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("swls_t2 ~ swls_t1", "d_swls_1 =~ swls_t2", "swls_t1 ~~ d_swls_1", # change parameters
"d_swls_1 ~1 ", "swls_t1 ~1 ", "", # means
"d_swls_1 ~~ d_swls_1")) # variances
kable(params_lcs_swls_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| swls_t2 ~ swls_t1 | 1.000 | 1.000 | 1.000 | 1.006 | NA | NA |
| d_swls_1 =~ swls_t2 | 1.000 | 1.000 | 1.000 | 0.566 | NA | NA |
| d_swls_1 ~1 | 0.156 | 0.072 | 0.239 | 0.275 | 4.775 | 0 |
| swls_t1 ~1 | 2.751 | 2.618 | 2.884 | 2.739 | 53.268 | 0 |
| d_swls_1 ~~ d_swls_1 | 0.320 | 0.189 | 0.450 | 1.000 | 6.308 | 0 |
| swls_t1 ~~ d_swls_1 | -0.166 | -0.255 | -0.076 | -0.292 | -4.784 | 0 |
Plot model:
semPaths(fit_mi_lcs_swls_hyp2, what="est",
sizeLat = 7, sizeMan = 7, edge.label.cex = .75, intercepts = T, edge.color="black")
Fit model:
# Code snippets adapted from Kievit et al. (2018) -- CC-BY -- https://doi.org/10.1016/j.dcn.2017.11.007
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_meaning_hyp2 <- '
meaning_t1 =~ 1*meaning_par1_t1 + lamb2*meaning_par2_t1 + lamb3*meaning_par3_t1 # This specifies the measurement model for meaning_t1
meaning_t2 =~ 1*meaning_par1_t2 + lamb2*meaning_par2_t2 + lamb3*meaning_par3_t2 # This specifies the measurement model for meaning_t2 with the equality constrained factor loadings
meaning_t2 ~ 1*meaning_t1 # This parameter regresses meaning_t2 perfectly on meaning_t1
d_meaning_1 =~ 1*meaning_t2 # This defines the latent change score factor as measured perfectly by scores on meaning_t2
meaning_t2 ~ 0*1 # This line constrains the intercept of meaning_t2 to 0
meaning_t2 ~~ 0*meaning_t2 # This fixes the variance of meaning_t2 to 0
d_meaning_1 ~ 1 # This estimates the intercept of the change score
meaning_t1 ~ 1 # This estimates the intercept of meaning_t1
d_meaning_1 ~~ d_meaning_1 # This estimates the variance of the change scores
meaning_t1 ~~ meaning_t1 # This estimates the variance of the meaning_t1
d_meaning_1 ~~ meaning_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
meaning_par1_t1 ~~ meaning_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
meaning_par2_t1 ~~ meaning_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
meaning_par3_t1 ~~ meaning_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
meaning_par1_t1 ~~ res1*meaning_par1_t1 # This allows residual variance on indicator X1 at T1
meaning_par2_t1 ~~ res2*meaning_par2_t1 # This allows residual variance on indicator X2 at T1
meaning_par3_t1 ~~ res3*meaning_par3_t1 # This allows residual variance on indicator X3 at T1
meaning_par1_t2 ~~ res1*meaning_par1_t2 # This allows residual variance on indicator X1 at T2
meaning_par2_t2 ~~ res2*meaning_par2_t2 # This allows residual variance on indicator X2 at T2
meaning_par3_t2 ~~ res3*meaning_par3_t2 # This allows residual variance on indicator X3 at T2
meaning_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
meaning_par2_t1 ~ m2*1 # This estimates the intercept of X2 at T1
meaning_par3_t1 ~ m3*1 # This estimates the intercept of X3 at T1
meaning_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
meaning_par2_t2 ~ m2*1 # This estimates the intercept of X2 at T2
meaning_par3_t2 ~ m3*1 # This estimates the intercept of X3 at T2
'
fit_mi_lcs_meaning_hyp2 <- lavaan(mi_lcs_meaning_hyp2, data=df_sbsa3_wide_wb, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_meaning_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_meaning_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 15 | 7541.413 | 7605.506 | 0.998 | 0.025 | 0.021 |
# parameters of interest
params_lcs_meaning_hyp2 <- broom::tidy(fit_mi_lcs_meaning_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("meaning_t2 ~ meaning_t1", "d_meaning_1 =~ meaning_t2", "meaning_t1 ~~ d_meaning_1", # change parameters
"d_meaning_1 ~1 ", "meaning_t1 ~1 ", "", # means
"d_meaning_1 ~~ d_meaning_1")) # variances
kable(params_lcs_meaning_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| meaning_t2 ~ meaning_t1 | 1.000 | 1.000 | 1.000 | 0.999 | NA | NA |
| d_meaning_1 =~ meaning_t2 | 1.000 | 1.000 | 1.000 | 0.578 | NA | NA |
| d_meaning_1 ~1 | 0.100 | 0.019 | 0.181 | 0.176 | 3.175 | 0.001 |
| meaning_t1 ~1 | 3.428 | 3.291 | 3.565 | 3.479 | 64.590 | 0.000 |
| d_meaning_1 ~~ d_meaning_1 | 0.325 | 0.214 | 0.436 | 1.000 | 7.526 | 0.000 |
| meaning_t1 ~~ d_meaning_1 | -0.161 | -0.263 | -0.060 | -0.287 | -4.087 | 0.000 |
Fit model:
# Code snippets adapted from Kievit et al. (2018) -- CC-BY -- https://doi.org/10.1016/j.dcn.2017.11.007
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_selfes_hyp2 <- '
selfes_t1 =~ 1*selfes_par1_t1 + lamb2*selfes_par2_t1 + lamb3*selfes_par3_t1 # This specifies the measurement model for selfes_t1
selfes_t2 =~ 1*selfes_par1_t2 + lamb2*selfes_par2_t2 + lamb3*selfes_par3_t2 # This specifies the measurement model for selfes_t2 with the equality constrained factor loadings
selfes_t2 ~ 1*selfes_t1 # This parameter regresses selfes_t2 perfectly on selfes_t1
d_selfes_1 =~ 1*selfes_t2 # This defines the latent change score factor as measured perfectly by scores on selfes_t2
selfes_t2 ~ 0*1 # This line constrains the intercept of selfes_t2 to 0
selfes_t2 ~~ 0*selfes_t2 # This fixes the variance of selfes_t2 to 0
d_selfes_1 ~ 1 # This estimates the intercept of the change score
selfes_t1 ~ 1 # This estimates the intercept of selfes_t1
d_selfes_1 ~~ d_selfes_1 # This estimates the variance of the change scores
selfes_t1 ~~ selfes_t1 # This estimates the variance of the selfes_t1
d_selfes_1 ~~ selfes_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
selfes_par1_t1 ~~ selfes_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
selfes_par2_t1 ~~ selfes_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
selfes_par3_t1 ~~ selfes_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
selfes_par1_t1 ~~ res1*selfes_par1_t1 # This allows residual variance on indicator X1 at T1
selfes_par2_t1 ~~ res2*selfes_par2_t1 # This allows residual variance on indicator X2 at T1
selfes_par3_t1 ~~ res3*selfes_par3_t1 # This allows residual variance on indicator X3 at T1
selfes_par1_t2 ~~ res1*selfes_par1_t2 # This allows residual variance on indicator X1 at T2
selfes_par2_t2 ~~ res2*selfes_par2_t2 # This allows residual variance on indicator X2 at T2
selfes_par3_t2 ~~ res3*selfes_par3_t2 # This allows residual variance on indicator X3 at T2
selfes_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
selfes_par2_t1 ~ m2*1 # This estimates the intercept of X2 at T1
selfes_par3_t1 ~ m3*1 # This estimates the intercept of X3 at T1
selfes_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
selfes_par2_t2 ~ m2*1 # This estimates the intercept of X2 at T2
selfes_par3_t2 ~ m3*1 # This estimates the intercept of X3 at T2
'
fit_mi_lcs_selfes_hyp2 <- lavaan(mi_lcs_selfes_hyp2, data=df_sbsa3_wide_wb, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_selfes_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_selfes_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 15 | 5506.131 | 5570.224 | 1 | 0 | 0.012 |
# parameters of interest
params_lcs_selfes_hyp2 <- broom::tidy(fit_mi_lcs_selfes_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("selfes_t2 ~ selfes_t1", "d_selfes_1 =~ selfes_t2", "selfes_t1 ~~ d_selfes_1", # change parameters
"d_selfes_1 ~1 ", "selfes_t1 ~1 ", "", # means
"d_selfes_1 ~~ d_selfes_1")) # variances
kable(params_lcs_selfes_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| selfes_t2 ~ selfes_t1 | 1.000 | 1.000 | 1.000 | 0.985 | NA | NA |
| d_selfes_1 =~ selfes_t2 | 1.000 | 1.000 | 1.000 | 0.505 | NA | NA |
| d_selfes_1 ~1 | 0.093 | 0.037 | 0.149 | 0.234 | 4.292 | 0 |
| selfes_t1 ~1 | 3.600 | 3.503 | 3.697 | 4.619 | 95.709 | 0 |
| d_selfes_1 ~~ d_selfes_1 | 0.160 | 0.107 | 0.213 | 1.000 | 7.840 | 0 |
| selfes_t1 ~~ d_selfes_1 | -0.070 | -0.115 | -0.026 | -0.226 | -4.067 | 0 |
Fit model:
# Code snippets adapted from Kievit et al. (2018) -- CC-BY -- https://doi.org/10.1016/j.dcn.2017.11.007
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_concept_hyp2 <- '
concept_t1 =~ 1*concept_par1_t1 + lamb2*concept_par2_t1 + lamb3*concept_par3_t1 # This specifies the measurement model for concept_t1
concept_t2 =~ 1*concept_par1_t2 + lamb2*concept_par2_t2 + lamb3*concept_par3_t2 # This specifies the measurement model for concept_t2 with the equality constrained factor loadings
concept_t2 ~ 1*concept_t1 # This parameter regresses concept_t2 perfectly on concept_t1
d_concept_1 =~ 1*concept_t2 # This defines the latent change score factor as measured perfectly by scores on concept_t2
concept_t2 ~ 0*1 # This line constrains the intercept of concept_t2 to 0
concept_t2 ~~ 0*concept_t2 # This fixes the variance of concept_t2 to 0
d_concept_1 ~ 1 # This estimates the intercept of the change score
concept_t1 ~ 1 # This estimates the intercept of concept_t1
d_concept_1 ~~ d_concept_1 # This estimates the variance of the change scores
concept_t1 ~~ concept_t1 # This estimates the variance of the concept_t1
d_concept_1 ~~ concept_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
concept_par1_t1 ~~ concept_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
concept_par2_t1 ~~ concept_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
concept_par3_t1 ~~ concept_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
concept_par1_t1 ~~ res1*concept_par1_t1 # This allows residual variance on indicator X1 at T1
concept_par2_t1 ~~ res2*concept_par2_t1 # This allows residual variance on indicator X2 at T1
concept_par3_t1 ~~ res3*concept_par3_t1 # This allows residual variance on indicator X3 at T1
concept_par1_t2 ~~ res1*concept_par1_t2 # This allows residual variance on indicator X1 at T2
concept_par2_t2 ~~ res2*concept_par2_t2 # This allows residual variance on indicator X2 at T2
concept_par3_t2 ~~ res3*concept_par3_t2 # This allows residual variance on indicator X3 at T2
concept_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
concept_par2_t1 ~ m2*1 # This estimates the intercept of X2 at T1
concept_par3_t1 ~ m3*1 # This estimates the intercept of X3 at T1
concept_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
concept_par2_t2 ~ m2*1 # This estimates the intercept of X2 at T2
concept_par3_t2 ~ m3*1 # This estimates the intercept of X3 at T2
'
fit_mi_lcs_concept_hyp2 <- lavaan(mi_lcs_concept_hyp2, data=df_sbsa3_wide_wb, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_concept_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_concept_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 15 | 5825.619 | 5889.713 | 0.997 | 0.033 | 0.022 |
# parameters of interest
params_lcs_concept_hyp2 <- broom::tidy(fit_mi_lcs_concept_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("concept_t2 ~ concept_t1", "d_concept_1 =~ concept_t2", "concept_t1 ~~ d_concept_1", # change parameters
"d_concept_1 ~1 ", "concept_t1 ~1 ", "", # means
"d_concept_1 ~~ d_concept_1")) # variances
kable(params_lcs_concept_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| concept_t2 ~ concept_t1 | 1.000 | 1.000 | 1.000 | 0.986 | NA | NA |
| d_concept_1 =~ concept_t2 | 1.000 | 1.000 | 1.000 | 0.593 | NA | NA |
| d_concept_1 ~1 | 0.105 | 0.044 | 0.166 | 0.236 | 4.428 | 0 |
| concept_t1 ~1 | 3.031 | 2.935 | 3.127 | 4.101 | 81.260 | 0 |
| d_concept_1 ~~ d_concept_1 | 0.198 | 0.135 | 0.260 | 1.000 | 8.160 | 0 |
| concept_t1 ~~ d_concept_1 | -0.091 | -0.138 | -0.044 | -0.278 | -5.019 | 0 |
The distance between ideal- and current personality will decrease in both groups.
We will use two strategies to test this hypothesis. First, we will compute the Fisher z-transformed profile correlation between current- and ideal personality and test whether it increased across assessments. Second, we will test whether the squared difference between current- and ideal personality ratings for each Big Five trait decreased across assessments. We will test mean-level differences in profile correlations and squared differences between baseline and follow up using repeated-measures t-test.
In Study 3, we can only test this in Group 1 where we assessed both current and ideal personality ratings.
Reshape to wide:
# reshape to wide
df_sbsa3_wide_profdiff <- df_sbsa3 %>%
filter(group=="Group 1") %>% # not possible in the other groups
arrange(pid, time) %>%
select(pid, time, profile_corr_item_z, profile_corr_facet_z, ends_with("_sqdiff")) %>%
pivot_wider(names_from = time,
names_sep = "_t",
values_from = -c(pid, time))Checking assumptions
# item-level profile correlation
shapiro.test(df_sbsa3_wide_profdiff$profile_corr_item_z_t2 - df_sbsa3_wide_profdiff$profile_corr_item_z_t1)
Shapiro-Wilk normality test
data: df_sbsa3_wide_profdiff$profile_corr_item_z_t2 - df_sbsa3_wide_profdiff$profile_corr_item_z_t1
W = 0.81981, p-value = 1.711e-12lillie.test(df_sbsa3_wide_profdiff$profile_corr_item_z_t2 - df_sbsa3_wide_profdiff$profile_corr_item_z_t1)
Lilliefors (Kolmogorov-Smirnov) normality test
data: df_sbsa3_wide_profdiff$profile_corr_item_z_t2 - df_sbsa3_wide_profdiff$profile_corr_item_z_t1
D = 0.11182, p-value = 6.975e-05# descriptives
psych::describe(df_sbsa3_wide_profdiff$profile_corr_item_z_t1) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 174 0.45 0.48 0.37 0.42 0.48 -0.62 1.91 2.53 0.56 0.33 0.04psych::describe(df_sbsa3_wide_profdiff$profile_corr_item_z_t2) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 155 0.54 0.54 0.47 0.49 0.42 -0.36 3.16 3.52 1.43 4.14 0.04# facet-level profile correlation
shapiro.test(df_sbsa3_wide_profdiff$profile_corr_facet_z_t2 - df_sbsa3_wide_profdiff$profile_corr_facet_z_t1)
Shapiro-Wilk normality test
data: df_sbsa3_wide_profdiff$profile_corr_facet_z_t2 - df_sbsa3_wide_profdiff$profile_corr_facet_z_t1
W = 0.94078, p-value = 4.647e-06lillie.test(df_sbsa3_wide_profdiff$profile_corr_facet_z_t2 - df_sbsa3_wide_profdiff$profile_corr_facet_z_t1)
Lilliefors (Kolmogorov-Smirnov) normality test
data: df_sbsa3_wide_profdiff$profile_corr_facet_z_t2 - df_sbsa3_wide_profdiff$profile_corr_facet_z_t1
D = 0.074446, p-value = 0.03642# descriptives
psych::describe(df_sbsa3_wide_profdiff$profile_corr_facet_z_t1) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 174 0.38 0.66 0.33 0.35 0.65 -0.78 2.36 3.14 0.45 -0.08 0.05psych::describe(df_sbsa3_wide_profdiff$profile_corr_facet_z_t2) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 155 0.48 0.69 0.39 0.43 0.61 -0.94 2.68 3.62 0.77 0.83 0.06Results summary:
t_profile_corr_item <- t.test(df_sbsa3_wide_profdiff$profile_corr_item_z_t2, df_sbsa3_wide_profdiff$profile_corr_item_z_t1, paired = TRUE)
t_profile_corr_facet <- t.test(df_sbsa3_wide_profdiff$profile_corr_facet_z_t2, df_sbsa3_wide_profdiff$profile_corr_facet_z_t1, paired = TRUE)
kable(bind_rows(tidy(t_profile_corr_item), tidy(t_profile_corr_facet)) %>% mutate(outcome = c("Item-level profile corr.", "Facet-level profile corr.")) %>% select(outcome, everything()), digits = 3)| outcome | estimate | statistic | p.value | parameter | conf.low | conf.high | method | alternative |
|---|---|---|---|---|---|---|---|---|
| Item-level profile corr. | 0.101 | 3.217 | 0.002 | 153 | 0.039 | 0.163 | Paired t-test | two.sided |
| Facet-level profile corr. | 0.117 | 3.099 | 0.002 | 153 | 0.042 | 0.191 | Paired t-test | two.sided |
# effect sizes
d_profile_corr_item <- psych::cohen.d(profile_corr_item_z ~ time, data = df_sbsa3 %>% filter(group=="Group 1"))
d_profile_corr_item$cohen.d lower effect upper
profile_corr_item_z -0.03676651 0.1802867 0.3970643d_profile_corr_facet <- psych::cohen.d(profile_corr_facet_z ~ time, data = df_sbsa3 %>% filter(group=="Group 1"))
d_profile_corr_facet$cohen.d lower effect upper
profile_corr_facet_z -0.05927102 0.1576606 0.3743511Plot
profile_df_plot <- df_sbsa3 %>%
filter(group=="Group 1") %>%
select(pid, time, profile_corr_item_z, profile_corr_facet_z) %>%
pivot_longer(-c(pid, time), names_to = "itemfacet", values_to = "corr") %>%
mutate(itemfacet2 = fct_recode(itemfacet, "Item-level" = "profile_corr_item_z", "Facet-level" = "profile_corr_facet_z"),
itemfacet2 = fct_reorder(itemfacet2, corr, .desc = F))
ggplot(profile_df_plot) +
aes(x = as.factor(time), y = corr) +
geom_boxplot() +
geom_violin(fill = NA) +
facet_wrap(vars(itemfacet2)) +
labs(x = "Measurement Occasion", y = "Profile correlation") + # title = "Distance between ideal- and current personality"
theme_bw()
Checking assumptions
# extra
shapiro.test(df_sbsa3_wide_profdiff$extra_sqdiff_t2 - df_sbsa3_wide_profdiff$extra_sqdiff_t1)
Shapiro-Wilk normality test
data: df_sbsa3_wide_profdiff$extra_sqdiff_t2 - df_sbsa3_wide_profdiff$extra_sqdiff_t1
W = 0.88604, p-value = 1.618e-09lillie.test(df_sbsa3_wide_profdiff$extra_sqdiff_t2 - df_sbsa3_wide_profdiff$extra_sqdiff_t1)
Lilliefors (Kolmogorov-Smirnov) normality test
data: df_sbsa3_wide_profdiff$extra_sqdiff_t2 - df_sbsa3_wide_profdiff$extra_sqdiff_t1
D = 0.13641, p-value = 2.122e-07# descriptives
psych::describe(df_sbsa3_wide_profdiff$extra_sqdiff_t1) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 174 1.75 2.01 1.17 1.4 1.37 0 14.06 14.06 2.48 9.26 0.15psych::describe(df_sbsa3_wide_profdiff$extra_sqdiff_t2) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 155 1.37 1.61 0.84 1.09 1.15 0 10.03 10.03 2.2 6.59 0.13# agree
shapiro.test(df_sbsa3_wide_profdiff$agree_sqdiff_t2 - df_sbsa3_wide_profdiff$agree_sqdiff_t1)
Shapiro-Wilk normality test
data: df_sbsa3_wide_profdiff$agree_sqdiff_t2 - df_sbsa3_wide_profdiff$agree_sqdiff_t1
W = 0.8918, p-value = 3.291e-09lillie.test(df_sbsa3_wide_profdiff$agree_sqdiff_t2 - df_sbsa3_wide_profdiff$agree_sqdiff_t1)
Lilliefors (Kolmogorov-Smirnov) normality test
data: df_sbsa3_wide_profdiff$agree_sqdiff_t2 - df_sbsa3_wide_profdiff$agree_sqdiff_t1
D = 0.16743, p-value = 2.129e-11# descriptives
psych::describe(df_sbsa3_wide_profdiff$agree_sqdiff_t1) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 174 0.4 0.6 0.14 0.26 0.2 0 3.67 3.67 2.51 7.37 0.05psych::describe(df_sbsa3_wide_profdiff$agree_sqdiff_t2) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 155 0.29 0.45 0.11 0.19 0.15 0 2.78 2.78 2.86 9.92 0.04# consc
shapiro.test(df_sbsa3_wide_profdiff$consc_sqdiff_t2 - df_sbsa3_wide_profdiff$consc_sqdiff_t1)
Shapiro-Wilk normality test
data: df_sbsa3_wide_profdiff$consc_sqdiff_t2 - df_sbsa3_wide_profdiff$consc_sqdiff_t1
W = 0.85898, p-value = 7.631e-11lillie.test(df_sbsa3_wide_profdiff$consc_sqdiff_t2 - df_sbsa3_wide_profdiff$consc_sqdiff_t1)
Lilliefors (Kolmogorov-Smirnov) normality test
data: df_sbsa3_wide_profdiff$consc_sqdiff_t2 - df_sbsa3_wide_profdiff$consc_sqdiff_t1
D = 0.17145, p-value = 5.525e-12# descriptives
psych::describe(df_sbsa3_wide_profdiff$consc_sqdiff_t1) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 174 1.8 2.35 1.09 1.3 1.36 0 13.44 13.44 2.23 5.46 0.18psych::describe(df_sbsa3_wide_profdiff$consc_sqdiff_t2) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 155 1.54 2.11 0.84 1.08 1.08 0 12.25 12.25 2.49 7 0.17# neuro
shapiro.test(df_sbsa3_wide_profdiff$neuro_sqdiff_t2 - df_sbsa3_wide_profdiff$neuro_sqdiff_t1)
Shapiro-Wilk normality test
data: df_sbsa3_wide_profdiff$neuro_sqdiff_t2 - df_sbsa3_wide_profdiff$neuro_sqdiff_t1
W = 0.97928, p-value = 0.02024lillie.test(df_sbsa3_wide_profdiff$neuro_sqdiff_t2 - df_sbsa3_wide_profdiff$neuro_sqdiff_t1)
Lilliefors (Kolmogorov-Smirnov) normality test
data: df_sbsa3_wide_profdiff$neuro_sqdiff_t2 - df_sbsa3_wide_profdiff$neuro_sqdiff_t1
D = 0.10473, p-value = 0.0002893# descriptives
psych::describe(df_sbsa3_wide_profdiff$neuro_sqdiff_t1) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 174 2.78 2.95 1.56 2.32 2.06 0 12.84 12.84 1.25 0.96 0.22psych::describe(df_sbsa3_wide_profdiff$neuro_sqdiff_t2) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 155 2.52 2.72 1.78 2.04 2.35 0 14.06 14.06 1.58 2.5 0.22# openn
shapiro.test(df_sbsa3_wide_profdiff$openn_sqdiff_t2 - df_sbsa3_wide_profdiff$openn_sqdiff_t1)
Shapiro-Wilk normality test
data: df_sbsa3_wide_profdiff$openn_sqdiff_t2 - df_sbsa3_wide_profdiff$openn_sqdiff_t1
W = 0.79025, p-value = 1.397e-13lillie.test(df_sbsa3_wide_profdiff$openn_sqdiff_t2 - df_sbsa3_wide_profdiff$openn_sqdiff_t1)
Lilliefors (Kolmogorov-Smirnov) normality test
data: df_sbsa3_wide_profdiff$openn_sqdiff_t2 - df_sbsa3_wide_profdiff$openn_sqdiff_t1
D = 0.18609, p-value = 3.017e-14# descriptives
psych::describe(df_sbsa3_wide_profdiff$openn_sqdiff_t1) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 174 0.4 0.74 0.11 0.24 0.15 0 6.67 6.67 4.41 29.21 0.06psych::describe(df_sbsa3_wide_profdiff$openn_sqdiff_t2) vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 155 0.34 0.5 0.17 0.23 0.25 0 3.06 3.06 2.81 9.45 0.04Results summary:
mod_traits_sqdiff <- df_sbsa3_wide_profdiff %>%
select(-starts_with("profile")) %>%
pivot_longer(-c(pid),
names_to = c("test", "time"), values_to = "score",
names_pattern = "(.*)_(t1|t2)") %>%
filter(!is.na(score)) %>%
group_by(pid, test) %>%
mutate(assessments = n()) %>%
ungroup() %>%
filter(assessments==2) %>%
select(-assessments) %>%
group_nest(test) %>%
mutate(t_tests = map(data, ~t.test(score ~ relevel(as.factor(time), "t2"), # need to relevel this to get T2-T1 diff (not sure why)
data = .x, paired = TRUE))) %>%
pull(t_tests) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
mod_traits_unlist <- mod_traits_sqdiff[[1]] %>% tidy()
for (i in 2:length(mod_traits_sqdiff)) {
mod_traits_unlist <- bind_rows(mod_traits_unlist, mod_traits_sqdiff[[i]] %>% tidy())
}
kable(mod_traits_unlist %>% mutate(outcome = sort(names(b5_vars))) %>%
select(outcome, everything()) %>% arrange(factor(outcome, levels = (names(b5_vars)))), digits = 3) # order by BFI traits| outcome | estimate | statistic | p.value | parameter | conf.low | conf.high | method | alternative |
|---|---|---|---|---|---|---|---|---|
| extraversion | -0.457 | -4.400 | 0.000 | 153 | -0.662 | -0.252 | Paired t-test | two.sided |
| agreeableness | -0.102 | -2.543 | 0.012 | 153 | -0.181 | -0.023 | Paired t-test | two.sided |
| conscientiousness | -0.352 | -3.106 | 0.002 | 153 | -0.577 | -0.128 | Paired t-test | two.sided |
| neuroticism | -0.327 | -2.214 | 0.028 | 153 | -0.618 | -0.035 | Paired t-test | two.sided |
| openness | -0.076 | -1.632 | 0.105 | 153 | -0.168 | 0.016 | Paired t-test | two.sided |
| sociability | -0.769 | -4.161 | 0.000 | 153 | -1.135 | -0.404 | Paired t-test | two.sided |
| assertiveness | -0.358 | -2.581 | 0.011 | 153 | -0.631 | -0.084 | Paired t-test | two.sided |
| energy | -0.279 | -1.739 | 0.084 | 153 | -0.596 | 0.038 | Paired t-test | two.sided |
| compassion | 0.070 | 1.017 | 0.311 | 153 | -0.066 | 0.205 | Paired t-test | two.sided |
| respectfulness | -0.047 | -0.845 | 0.399 | 153 | -0.159 | 0.064 | Paired t-test | two.sided |
| trust | -0.384 | -2.876 | 0.005 | 153 | -0.647 | -0.120 | Paired t-test | two.sided |
| organization | -0.441 | -2.412 | 0.017 | 153 | -0.802 | -0.080 | Paired t-test | two.sided |
| productiveness | -0.673 | -3.139 | 0.002 | 153 | -1.096 | -0.249 | Paired t-test | two.sided |
| responsibility | -0.211 | -2.119 | 0.036 | 153 | -0.408 | -0.014 | Paired t-test | two.sided |
| anxiety | -0.390 | -1.524 | 0.130 | 153 | -0.895 | 0.116 | Paired t-test | two.sided |
| depression | -0.588 | -2.776 | 0.006 | 153 | -1.006 | -0.170 | Paired t-test | two.sided |
| volatility | -0.181 | -0.910 | 0.364 | 153 | -0.574 | 0.212 | Paired t-test | two.sided |
| curiosity | -0.103 | -1.400 | 0.164 | 153 | -0.249 | 0.042 | Paired t-test | two.sided |
| aesthetic | -0.041 | -0.450 | 0.654 | 153 | -0.219 | 0.138 | Paired t-test | two.sided |
| imagination | -0.146 | -1.281 | 0.202 | 153 | -0.370 | 0.079 | Paired t-test | two.sided |
# effect sizes
cohend_cust = function(outcome, df) {
form = as.formula(paste0(outcome, " ~ time"))
psych::cohen.d(form, data = df)
}
d_trait <- cohend_cust(paste0(str_trunc(names(b5_vars)[1], 5, ellipsis = ""), "_sqdiff"),
df_sbsa3 %>% filter(group=="Group 1"))
d_traits_unlist <- as_tibble(d_trait$cohen.d)
for (i in 2:length(names(b5_vars))) {
d_trait <- cohend_cust(paste0(str_trunc(names(b5_vars)[i], 5, ellipsis = ""), "_sqdiff"),
df_sbsa3 %>% filter(group=="Group 1"))
d_traits_unlist <- bind_rows(d_traits_unlist, as_tibble(d_trait$cohen.d))
}
kable(d_traits_unlist %>% mutate(outcome = names(b5_vars)) %>% select(outcome, everything()), digits = 3)| outcome | lower | effect | upper |
|---|---|---|---|
| extraversion | -0.424 | -0.207 | 0.010 |
| agreeableness | -0.427 | -0.210 | 0.007 |
| conscientiousness | -0.334 | -0.118 | 0.099 |
| neuroticism | -0.307 | -0.090 | 0.127 |
| openness | -0.316 | -0.099 | 0.117 |
| sociability | -0.465 | -0.248 | -0.030 |
| assertiveness | -0.341 | -0.125 | 0.092 |
| energy | -0.307 | -0.090 | 0.126 |
| compassion | -0.118 | 0.099 | 0.315 |
| respectfulness | -0.292 | -0.076 | 0.141 |
| trust | -0.461 | -0.244 | -0.026 |
| organization | -0.328 | -0.112 | 0.105 |
| productiveness | -0.394 | -0.178 | 0.040 |
| responsibility | -0.291 | -0.074 | 0.142 |
| anxiety | -0.289 | -0.073 | 0.144 |
| depression | -0.350 | -0.133 | 0.084 |
| volatility | -0.264 | -0.048 | 0.169 |
| curiosity | -0.293 | -0.077 | 0.140 |
| aesthetic | -0.262 | -0.046 | 0.171 |
| imagination | -0.280 | -0.063 | 0.153 |
[Note: These CIs for d are probably “wrong” -> not appropriate for paired t-test]
Plot (traits/facets sorted by average size of squared differences)
sqdiff_df_plot <- df_sbsa3 %>%
filter(group=="Group 1") %>%
select(pid, time, all_of(paste0(str_trunc(names(b5_vars), 5, ellipsis = ""), "_sqdiff"))) %>%
pivot_longer(-c(pid, time), names_to = "trait", values_to = "sqdiff") %>%
group_by(trait) %>%
mutate(cur_group_id()) %>%
ungroup() %>%
mutate(trait2 = fct_recode(trait, !!! setNames(paste0(str_trunc(names(b5_vars), 5, ellipsis = ""), "_sqdiff"),
str_to_title(names(b5_vars)))),
trait2 = fct_reorder(trait2, sqdiff, .desc = T))
plot_sqdiff <- ggplot(sqdiff_df_plot) +
aes(x = as.factor(time), y = sqdiff) +
geom_boxplot() +
geom_violin(fill = NA) +
ggforce::facet_wrap_paginate(~trait2, ncol=3, nrow=2) +
labs(x = "Measurement Occasion", y = "Squared difference") +
theme_bw()
plot_sqdiff + ggforce::facet_wrap_paginate(~trait2, ncol=4, nrow=1, page = 1)
plot_sqdiff + ggforce::facet_wrap_paginate(~trait2, ncol=4, nrow=1, page = 2)
plot_sqdiff + ggforce::facet_wrap_paginate(~trait2, ncol=4, nrow=1, page = 3)
plot_sqdiff + ggforce::facet_wrap_paginate(~trait2, ncol=4, nrow=1, page = 4)
plot_sqdiff + ggforce::facet_wrap_paginate(~trait2, ncol=4, nrow=1, page = 5)
Amount of change on the latent level collapsed across all groups.
We will test the mean-level difference between baseline and follow up using a latent change model.
Fit model:
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_extra_hyp2 <- '
extra_t1 =~ 1*extra_curr_par1_t1 + lamb2*extra_curr_par2_t1 + lamb3*extra_curr_par3_t1 # Measurement model T1
extra_t2 =~ 1*extra_curr_par1_t2 + lamb2*extra_curr_par2_t2 + lamb3*extra_curr_par3_t2 # Measurement model T2 with the equality constrained factor loadings
extra_t2 ~ 1*extra_t1 # This parameter regresses extra_t2 perfectly on extra_t1
d_extra_1 =~ 1*extra_t2 # This defines the latent change score factor as measured perfectly by scores on extra_t2
extra_t2 ~ 0*1 # This line constrains the intercept of extra_t2 to 0
extra_t2 ~~ 0*extra_t2 # This fixes the variance of extra_t2 to 0
d_extra_1 ~ 1 # This estimates the intercept of the change score
extra_t1 ~ 1 # This estimates the intercept of extra_t1
d_extra_1 ~~ d_extra_1 # This estimates the variance of the change scores
extra_t1 ~~ extra_t1 # This estimates the variance of the extra_t1
d_extra_1 ~~ extra_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
extra_curr_par1_t1 ~~ extra_curr_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
extra_curr_par2_t1 ~~ extra_curr_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
extra_curr_par3_t1 ~~ extra_curr_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
extra_curr_par1_t1 ~~ res1*extra_curr_par1_t1 # This allows residual variance on indicator X1 at T1
extra_curr_par2_t1 ~~ res2*extra_curr_par2_t1 # This allows residual variance on indicator X2 at T1
extra_curr_par3_t1 ~~ res3*extra_curr_par3_t1 # This allows residual variance on indicator X3 at T1
extra_curr_par1_t2 ~~ res1*extra_curr_par1_t2 # This allows residual variance on indicator X1 at T2
extra_curr_par2_t2 ~~ res2*extra_curr_par2_t2 # This allows residual variance on indicator X2 at T2
extra_curr_par3_t2 ~~ res3*extra_curr_par3_t2 # This allows residual variance on indicator X3 at T2
extra_curr_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
extra_curr_par2_t1 ~ m2*1 # This estimates the intercept of X2 at T1
extra_curr_par3_t1 ~ m3*1 # This estimates the intercept of X3 at T1
extra_curr_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
extra_curr_par2_t2 ~ m2*1 # This estimates the intercept of X2 at T2
extra_curr_par3_t2 ~ m3*1 # This estimates the intercept of X3 at T2
'
fit_mi_lcs_extra_hyp2 <- lavaan(mi_lcs_extra_hyp2, data=df_sbsa3_wide_pers, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_extra_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_extra_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 15 | 4920.539 | 4984.632 | 1 | 0 | 0.017 |
# parameters of interest
params_lcs_extra_hyp2 <- broom::tidy(fit_mi_lcs_extra_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("extra_t2 ~ extra_t1", "d_extra_1 =~ extra_t2", "extra_t1 ~~ d_extra_1", # change parameters
"d_extra_1 ~1 ", "extra_t1 ~1 ", "", # means
"d_extra_1 ~~ d_extra_1")) # variances
kable(params_lcs_extra_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| extra_t2 ~ extra_t1 | 1.000 | 1.000 | 1.000 | 0.989 | NA | NA |
| d_extra_1 =~ extra_t2 | 1.000 | 1.000 | 1.000 | 0.329 | NA | NA |
| d_extra_1 ~1 | 0.043 | 0.000 | 0.085 | 0.180 | 2.562 | 0.01 |
| extra_t1 ~1 | 3.078 | 2.987 | 3.170 | 4.314 | 86.648 | 0.00 |
| d_extra_1 ~~ d_extra_1 | 0.056 | 0.026 | 0.086 | 1.000 | 4.835 | 0.00 |
| extra_t1 ~~ d_extra_1 | -0.023 | -0.059 | 0.014 | -0.134 | -1.597 | 0.11 |
Fit model:
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_agree_hyp2 <- '
agree_t1 =~ 1*agree_curr_par1_t1 + lamb2*agree_curr_par2_t1 + lamb3*agree_curr_par3_t1 # Measurement model T1
agree_t2 =~ 1*agree_curr_par1_t2 + lamb2*agree_curr_par2_t2 + lamb3*agree_curr_par3_t2 # Measurement model T2 with the equality constrained factor loadings
agree_t2 ~ 1*agree_t1 # This parameter regresses agree_t2 perfectly on agree_t1
d_agree_1 =~ 1*agree_t2 # This defines the latent change score factor as measured perfectly by scores on agree_t2
agree_t2 ~ 0*1 # This line constrains the intercept of agree_t2 to 0
agree_t2 ~~ 0*agree_t2 # This fixes the variance of agree_t2 to 0
d_agree_1 ~ 1 # This estimates the intercept of the change score
agree_t1 ~ 1 # This estimates the intercept of agree_t1
d_agree_1 ~~ d_agree_1 # This estimates the variance of the change scores
agree_t1 ~~ agree_t1 # This estimates the variance of the agree_t1
d_agree_1 ~~ agree_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
agree_curr_par1_t1 ~~ agree_curr_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
agree_curr_par2_t1 ~~ agree_curr_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
agree_curr_par3_t1 ~~ agree_curr_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
agree_curr_par1_t1 ~~ res1*agree_curr_par1_t1 # This allows residual variance on indicator X1 at T1
agree_curr_par2_t1 ~~ res2*agree_curr_par2_t1 # This allows residual variance on indicator X2 at T1
agree_curr_par3_t1 ~~ res3*agree_curr_par3_t1 # This allows residual variance on indicator X3 at T1
agree_curr_par1_t2 ~~ res1*agree_curr_par1_t2 # This allows residual variance on indicator X1 at T2
agree_curr_par2_t2 ~~ res2*agree_curr_par2_t2 # This allows residual variance on indicator X2 at T2
agree_curr_par3_t2 ~~ res3*agree_curr_par3_t2 # This allows residual variance on indicator X3 at T2
agree_curr_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
agree_curr_par2_t1 ~ m2*1 # This estimates the intercept of X2 at T1
agree_curr_par3_t1 ~ m3*1 # This estimates the intercept of X3 at T1
agree_curr_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
agree_curr_par2_t2 ~ m2*1 # This estimates the intercept of X2 at T2
agree_curr_par3_t2 ~ m3*1 # This estimates the intercept of X3 at T2
'
fit_mi_lcs_agree_hyp2 <- lavaan(mi_lcs_agree_hyp2, data=df_sbsa3_wide_pers, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_agree_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_agree_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 15 | 4787.496 | 4851.589 | 0.987 | 0.057 | 0.03 |
# parameters of interest
params_lcs_agree_hyp2 <- broom::tidy(fit_mi_lcs_agree_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("agree_t2 ~ agree_t1", "d_agree_1 =~ agree_t2", "agree_t1 ~~ d_agree_1", # change parameters
"d_agree_1 ~1 ", "agree_t1 ~1 ", "", # means
"d_agree_1 ~~ d_agree_1")) # variances
kable(params_lcs_agree_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| agree_t2 ~ agree_t1 | 1.000 | 1.000 | 1.000 | 0.999 | NA | NA |
| d_agree_1 =~ agree_t2 | 1.000 | 1.000 | 1.000 | 0.502 | NA | NA |
| d_agree_1 ~1 | 0.026 | -0.014 | 0.066 | 0.110 | 1.687 | 0.092 |
| agree_t1 ~1 | 3.732 | 3.657 | 3.807 | 7.943 | 127.758 | 0.000 |
| d_agree_1 ~~ d_agree_1 | 0.056 | 0.018 | 0.094 | 1.000 | 3.789 | 0.000 |
| agree_t1 ~~ d_agree_1 | -0.028 | -0.052 | -0.003 | -0.249 | -2.937 | 0.003 |
Fit model:
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_consc_hyp2 <- '
consc_t1 =~ 1*consc_curr_par1_t1 + lamb2*consc_curr_par2_t1 + lamb3*consc_curr_par3_t1 # Measurement model T1
consc_t2 =~ 1*consc_curr_par1_t2 + lamb2*consc_curr_par2_t2 + lamb3*consc_curr_par3_t2 # Measurement model T2 with the equality constrained factor loadings
consc_t2 ~ 1*consc_t1 # This parameter regresses consc_t2 perfectly on consc_t1
d_consc_1 =~ 1*consc_t2 # This defines the latent change score factor as measured perfectly by scores on consc_t2
consc_t2 ~ 0*1 # This line constrains the intercept of consc_t2 to 0
consc_t2 ~~ 0*consc_t2 # This fixes the variance of consc_t2 to 0
d_consc_1 ~ 1 # This estimates the intercept of the change score
consc_t1 ~ 1 # This estimates the intercept of consc_t1
d_consc_1 ~~ d_consc_1 # This estimates the variance of the change scores
consc_t1 ~~ consc_t1 # This estimates the variance of the consc_t1
d_consc_1 ~~ consc_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
consc_curr_par1_t1 ~~ consc_curr_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
consc_curr_par2_t1 ~~ consc_curr_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
consc_curr_par3_t1 ~~ consc_curr_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
consc_curr_par1_t1 ~~ res1*consc_curr_par1_t1 # This allows residual variance on indicator X1 at T1
consc_curr_par2_t1 ~~ res2*consc_curr_par2_t1 # This allows residual variance on indicator X2 at T1
consc_curr_par3_t1 ~~ res3*consc_curr_par3_t1 # This allows residual variance on indicator X3 at T1
consc_curr_par1_t2 ~~ res1*consc_curr_par1_t2 # This allows residual variance on indicator X1 at T2
consc_curr_par2_t2 ~~ res2*consc_curr_par2_t2 # This allows residual variance on indicator X2 at T2
consc_curr_par3_t2 ~~ res3*consc_curr_par3_t2 # This allows residual variance on indicator X3 at T2
consc_curr_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
consc_curr_par2_t1 ~ m2*1 # This estimates the intercept of X2 at T1
consc_curr_par3_t1 ~ m3*1 # This estimates the intercept of X3 at T1
consc_curr_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
consc_curr_par2_t2 ~ m2*1 # This estimates the intercept of X2 at T2
consc_curr_par3_t2 ~ m3*1 # This estimates the intercept of X3 at T2
'
fit_mi_lcs_consc_hyp2 <- lavaan(mi_lcs_consc_hyp2, data=df_sbsa3_wide_pers, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_consc_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_consc_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 15 | 4835.468 | 4899.561 | 0.998 | 0.024 | 0.013 |
# parameters of interest
params_lcs_consc_hyp2 <- broom::tidy(fit_mi_lcs_consc_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("consc_t2 ~ consc_t1", "d_consc_1 =~ consc_t2", "consc_t1 ~~ d_consc_1", # change parameters
"d_consc_1 ~1 ", "consc_t1 ~1 ", "", # means
"d_consc_1 ~~ d_consc_1")) # variances
kable(params_lcs_consc_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| consc_t2 ~ consc_t1 | 1.000 | 1.000 | 1.000 | 1.029 | NA | NA |
| d_consc_1 =~ consc_t2 | 1.000 | 1.000 | 1.000 | 0.386 | NA | NA |
| d_consc_1 ~1 | 0.068 | 0.023 | 0.112 | 0.255 | 3.899 | 0 |
| consc_t1 ~1 | 3.161 | 3.070 | 3.252 | 4.467 | 89.194 | 0 |
| d_consc_1 ~~ d_consc_1 | 0.070 | 0.038 | 0.103 | 1.000 | 5.535 | 0 |
| consc_t1 ~~ d_consc_1 | -0.049 | -0.085 | -0.014 | -0.262 | -3.577 | 0 |
Fit model:
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_neuro_hyp2 <- '
neuro_t1 =~ 1*neuro_curr_par1_t1 + lamb2*neuro_curr_par2_t1 + lamb3*neuro_curr_par3_t1 # Measurement model T1
neuro_t2 =~ 1*neuro_curr_par1_t2 + lamb2*neuro_curr_par2_t2 + lamb3*neuro_curr_par3_t2 # Measurement model T2 with the equality constrained factor loadings
neuro_t2 ~ 1*neuro_t1 # This parameter regresses neuro_t2 perfectly on neuro_t1
d_neuro_1 =~ 1*neuro_t2 # This defines the latent change score factor as measured perfectly by scores on neuro_t2
neuro_t2 ~ 0*1 # This line constrains the intercept of neuro_t2 to 0
neuro_t2 ~~ 0*neuro_t2 # This fixes the variance of neuro_t2 to 0
d_neuro_1 ~ 1 # This estimates the intercept of the change score
neuro_t1 ~ 1 # This estimates the intercept of neuro_t1
d_neuro_1 ~~ d_neuro_1 # This estimates the variance of the change scores
neuro_t1 ~~ neuro_t1 # This estimates the variance of the neuro_t1
d_neuro_1 ~~ neuro_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
neuro_curr_par1_t1 ~~ neuro_curr_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
neuro_curr_par2_t1 ~~ neuro_curr_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
neuro_curr_par3_t1 ~~ neuro_curr_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
neuro_curr_par1_t1 ~~ res1*neuro_curr_par1_t1 # This allows residual variance on indicator X1 at T1
neuro_curr_par2_t1 ~~ res2*neuro_curr_par2_t1 # This allows residual variance on indicator X2 at T1
neuro_curr_par3_t1 ~~ res3*neuro_curr_par3_t1 # This allows residual variance on indicator X3 at T1
neuro_curr_par1_t2 ~~ res1*neuro_curr_par1_t2 # This allows residual variance on indicator X1 at T2
neuro_curr_par2_t2 ~~ res2*neuro_curr_par2_t2 # This allows residual variance on indicator X2 at T2
neuro_curr_par3_t2 ~~ res3*neuro_curr_par3_t2 # This allows residual variance on indicator X3 at T2
neuro_curr_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
neuro_curr_par2_t1 ~ m2*1 # This estimates the intercept of X2 at T1
neuro_curr_par3_t1 ~ m3*1 # This estimates the intercept of X3 at T1
neuro_curr_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
neuro_curr_par2_t2 ~ m2*1 # This estimates the intercept of X2 at T2
neuro_curr_par3_t2 ~ m3*1 # This estimates the intercept of X3 at T2
'
fit_mi_lcs_neuro_hyp2 <- lavaan(mi_lcs_neuro_hyp2, data=df_sbsa3_wide_pers, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_neuro_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_neuro_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 15 | 4921.769 | 4985.862 | 0.999 | 0.026 | 0.015 |
# parameters of interest
params_lcs_neuro_hyp2 <- broom::tidy(fit_mi_lcs_neuro_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("neuro_t2 ~ neuro_t1", "d_neuro_1 =~ neuro_t2", "neuro_t1 ~~ d_neuro_1", # change parameters
"d_neuro_1 ~1 ", "neuro_t1 ~1 ", "", # means
"d_neuro_1 ~~ d_neuro_1")) # variances
kable(params_lcs_neuro_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| neuro_t2 ~ neuro_t1 | 1.000 | 1.000 | 1.000 | 1.018 | NA | NA |
| d_neuro_1 =~ neuro_t2 | 1.000 | 1.000 | 1.000 | 0.400 | NA | NA |
| d_neuro_1 ~1 | -0.053 | -0.101 | -0.005 | -0.167 | -2.821 | 0.005 |
| neuro_t1 ~1 | 3.130 | 3.029 | 3.231 | 3.889 | 79.932 | 0.000 |
| d_neuro_1 ~~ d_neuro_1 | 0.100 | 0.062 | 0.139 | 1.000 | 6.698 | 0.000 |
| neuro_t1 ~~ d_neuro_1 | -0.061 | -0.102 | -0.021 | -0.241 | -3.887 | 0.000 |
Fit model:
# Fit the multiple indicator Univariate Latent Change Score model
mi_lcs_openn_hyp2 <- '
openn_t1 =~ 1*openn_curr_par1_t1 + lamb2*openn_curr_par2_t1 + lamb3*openn_curr_par3_t1 # Measurement model T1
openn_t2 =~ 1*openn_curr_par1_t2 + lamb2*openn_curr_par2_t2 + lamb3*openn_curr_par3_t2 # Measurement model T2 with the equality constrained factor loadings
openn_t2 ~ 1*openn_t1 # This parameter regresses openn_t2 perfectly on openn_t1
d_openn_1 =~ 1*openn_t2 # This defines the latent change score factor as measured perfectly by scores on openn_t2
openn_t2 ~ 0*1 # This line constrains the intercept of openn_t2 to 0
openn_t2 ~~ 0*openn_t2 # This fixes the variance of openn_t2 to 0
d_openn_1 ~ 1 # This estimates the intercept of the change score
openn_t1 ~ 1 # This estimates the intercept of openn_t1
d_openn_1 ~~ d_openn_1 # This estimates the variance of the change scores
openn_t1 ~~ openn_t1 # This estimates the variance of the openn_t1
d_openn_1 ~~ openn_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
openn_curr_par1_t1 ~~ openn_curr_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
openn_curr_par2_t1 ~~ openn_curr_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
openn_curr_par3_t1 ~~ openn_curr_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
openn_curr_par1_t1 ~~ res1*openn_curr_par1_t1 # This allows residual variance on indicator X1 at T1
openn_curr_par2_t1 ~~ res2*openn_curr_par2_t1 # This allows residual variance on indicator X2 at T1
openn_curr_par3_t1 ~~ res3*openn_curr_par3_t1 # This allows residual variance on indicator X3 at T1
openn_curr_par1_t2 ~~ res1*openn_curr_par1_t2 # This allows residual variance on indicator X1 at T2
openn_curr_par2_t2 ~~ res2*openn_curr_par2_t2 # This allows residual variance on indicator X2 at T2
openn_curr_par3_t2 ~~ res3*openn_curr_par3_t2 # This allows residual variance on indicator X3 at T2
openn_curr_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
openn_curr_par2_t1 ~ m2*1 # This estimates the intercept of X2 at T1
openn_curr_par3_t1 ~ m3*1 # This estimates the intercept of X3 at T1
openn_curr_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
openn_curr_par2_t2 ~ m2*1 # This estimates the intercept of X2 at T2
openn_curr_par3_t2 ~ m3*1 # This estimates the intercept of X3 at T2
'
fit_mi_lcs_openn_hyp2 <- lavaan(mi_lcs_openn_hyp2, data=df_sbsa3_wide_pers, estimator='mlr', fixed.x=FALSE, missing='fiml')
summary(fit_mi_lcs_openn_hyp2, fit.measures=TRUE, standardized=TRUE, rsquare=F)Results summary:
# model fit
kable(broom::glance(fit_mi_lcs_openn_hyp2) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 530 | 15 | 4997.663 | 5061.757 | 0.999 | 0.017 | 0.014 |
# parameters of interest
params_lcs_openn_hyp2 <- broom::tidy(fit_mi_lcs_openn_hyp2, conf.int = TRUE, conf.level = 0.99) %>%
select(term, estimate, conf.low, conf.high, std.all, statistic, p.value) %>%
filter(term %in% c("openn_t2 ~ openn_t1", "d_openn_1 =~ openn_t2", "openn_t1 ~~ d_openn_1", # change parameters
"d_openn_1 ~1 ", "openn_t1 ~1 ", "", # means
"d_openn_1 ~~ d_openn_1")) # variances
kable(params_lcs_openn_hyp2, digits = 3)| term | estimate | conf.low | conf.high | std.all | statistic | p.value |
|---|---|---|---|---|---|---|
| openn_t2 ~ openn_t1 | 1.000 | 1.000 | 1.000 | 1.014 | NA | NA |
| d_openn_1 =~ openn_t2 | 1.000 | 1.000 | 1.000 | 0.335 | NA | NA |
| d_openn_1 ~1 | 0.019 | -0.015 | 0.053 | 0.117 | 1.422 | 0.155 |
| openn_t1 ~1 | 3.482 | 3.399 | 3.565 | 7.209 | 108.489 | 0.000 |
| d_openn_1 ~~ d_openn_1 | 0.026 | 0.003 | 0.048 | 1.000 | 2.902 | 0.004 |
| openn_t1 ~~ d_openn_1 | -0.016 | -0.036 | 0.004 | -0.206 | -2.040 | 0.041 |
We will explore a) whether change in psychological well-being indicators as well as the difference between real- and ideal- self will differ across groups at follow-up.
Create ordered factors first:
df_sbsa3_wide_wb <- df_sbsa3_wide_wb %>%
mutate(group = factor(group, ordered = T)) %>%
arrange(group, pid)
df_sbsa3_wide_pers <- df_sbsa3_wide_pers %>%
mutate(group = factor(group, ordered = T)) %>%
arrange(group, pid)Life satisfaction: fitting multi-group models
# adapt latent change score model from above and add grouping factor in estimation (also add vectorized equality constraints to the model)
# configural invariance
mi_lcs_swls_group_config <- '
swls_t1 =~ 1*sw06_01_t1 + c("lamb2a", "lamb2b")*sw06_02_t1 + c("lamb3a", "lamb3b")*sw06_03_t1 + c("lamb4a", "lamb4b")*sw06_04_t1 # This specifies the measurement model for swls_t1
swls_t2 =~ 1*sw06_01_t2 + c("lamb2a", "lamb2b")*sw06_02_t2 + c("lamb3a", "lamb3b")*sw06_03_t2 + c("lamb4a", "lamb4b")*sw06_04_t2 # This specifies the measurement model for swls_t2 with the equality constrained factor loadings
swls_t2 ~ 1*swls_t1 # This parameter regresses swls_t2 perfectly on swls_t1
d_swls_1 =~ 1*swls_t2 # This defines the latent change score factor as measured perfectly by scores on swls_t2
swls_t2 ~ 0*1 # This line constrains the intercept of swls_t2 to 0
swls_t2 ~~ 0*swls_t2 # This fixes the variance of swls_t2 to 0
d_swls_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
swls_t1 ~ c("wb_int_a", "wb_int_b")*1 # This estimates the intercept of swls_t1
d_swls_1 ~~ c("d_var_a", "d_var_b")*d_swls_1 # This estimates the variance of the change scores
swls_t1 ~~ c("wb_var_a", "wb_var_b")*swls_t1 # This estimates the variance of the swls_t1
d_swls_1 ~~ c("fb_a", "fb_b")*swls_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score remains unconditional
sw06_01_t1 ~~ c("cov1a", "cov1b")*sw06_01_t2 # This allows residual covariance on indicator X1 across T1 and T2
sw06_02_t1 ~~ c("cov2a", "cov2b")*sw06_02_t2 # This allows residual covariance on indicator X2 across T1 and T2
sw06_03_t1 ~~ c("cov3a", "cov3b")*sw06_03_t2 # This allows residual covariance on indicator X3 across T1 and T2
sw06_04_t1 ~~ c("cov4a", "cov4b")*sw06_04_t2 # This allows residual covariance on indicator X4 across T1 and T2
sw06_01_t1 ~~ c("res1a", "res1b")*sw06_01_t1 # This allows residual variance on indicator X1 at T1
sw06_02_t1 ~~ c("res2a", "res2b")*sw06_02_t1 # This allows residual variance on indicator X2 at T1
sw06_03_t1 ~~ c("res3a", "res3b")*sw06_03_t1 # This allows residual variance on indicator X3 at T1
sw06_04_t1 ~~ c("res4a", "res4b")*sw06_04_t1 # This allows residual variance on indicator X4 at T1
sw06_01_t2 ~~ c("res1a", "res1b")*sw06_01_t2 # This allows residual variance on indicator X1 at T2
sw06_02_t2 ~~ c("res2a", "res2b")*sw06_02_t2 # This allows residual variance on indicator X2 at T2
sw06_03_t2 ~~ c("res3a", "res3b")*sw06_03_t2 # This allows residual variance on indicator X3 at T2
sw06_04_t2 ~~ c("res4a", "res4b")*sw06_04_t2 # This allows residual variance on indicator X4 at T2
sw06_01_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
sw06_02_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
sw06_03_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
sw06_04_t1 ~ c("m4a", "m4b")*1 # This estimates the intercept of X4 at T1
sw06_01_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
sw06_02_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
sw06_03_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
sw06_04_t2 ~ c("m4a", "m4b")*1 # This estimates the intercept of X4 at T2
'
lcs_swls_group_12_config <- sem(mi_lcs_swls_group_config, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr',
fixed.x=FALSE, missing='fiml', group = "group")
# weak invariance
mi_lcs_swls_group_weak <- '
swls_t1 =~ 1*sw06_01_t1 + c("lamb2", "lamb2")*sw06_02_t1 + c("lamb3", "lamb3")*sw06_03_t1 + c("lamb4", "lamb4")*sw06_04_t1 # This specifies the measurement model for swls_t1
swls_t2 =~ 1*sw06_01_t2 + c("lamb2", "lamb2")*sw06_02_t2 + c("lamb3", "lamb3")*sw06_03_t2 + c("lamb4", "lamb4")*sw06_04_t2 # This specifies the measurement model for swls_t2 with the equality constrained factor loadings
swls_t2 ~ 1*swls_t1 # This parameter regresses swls_t2 perfectly on swls_t1
d_swls_1 =~ 1*swls_t2 # This defines the latent change score factor as measured perfectly by scores on swls_t2
swls_t2 ~ 0*1 # This line constrains the intercept of swls_t2 to 0
swls_t2 ~~ 0*swls_t2 # This fixes the variance of swls_t2 to 0
d_swls_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
swls_t1 ~ c("wb_int_a", "wb_int_b")*1 # This estimates the intercept of swls_t1
d_swls_1 ~~ c("d_var_a", "d_var_b")*d_swls_1 # This estimates the variance of the change scores
swls_t1 ~~ c("wb_var_a", "wb_var_b")*swls_t1 # This estimates the variance of the swls_t1
d_swls_1 ~~ c("fb_a", "fb_b")*swls_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
sw06_01_t1 ~~ c("cov1a", "cov1b")*sw06_01_t2 # This allows residual covariance on indicator X1 across T1 and T2
sw06_02_t1 ~~ c("cov2a", "cov2b")*sw06_02_t2 # This allows residual covariance on indicator X2 across T1 and T2
sw06_03_t1 ~~ c("cov3a", "cov3b")*sw06_03_t2 # This allows residual covariance on indicator X3 across T1 and T2
sw06_04_t1 ~~ c("cov4a", "cov4b")*sw06_04_t2 # This allows residual covariance on indicator X4 across T1 and T2
sw06_01_t1 ~~ c("res1a", "res1b")*sw06_01_t1 # This allows residual variance on indicator X1 at T1
sw06_02_t1 ~~ c("res2a", "res2b")*sw06_02_t1 # This allows residual variance on indicator X2 at T1
sw06_03_t1 ~~ c("res3a", "res3b")*sw06_03_t1 # This allows residual variance on indicator X3 at T1
sw06_04_t1 ~~ c("res4a", "res4b")*sw06_04_t1 # This allows residual variance on indicator X4 at T1
sw06_01_t2 ~~ c("res1a", "res1b")*sw06_01_t2 # This allows residual variance on indicator X1 at T2
sw06_02_t2 ~~ c("res2a", "res2b")*sw06_02_t2 # This allows residual variance on indicator X2 at T2
sw06_03_t2 ~~ c("res3a", "res3b")*sw06_03_t2 # This allows residual variance on indicator X3 at T2
sw06_04_t2 ~~ c("res4a", "res4b")*sw06_04_t2 # This allows residual variance on indicator X4 at T2
sw06_01_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
sw06_02_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
sw06_03_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
sw06_04_t1 ~ c("m4a", "m4b")*1 # This estimates the intercept of X4 at T1
sw06_01_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
sw06_02_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
sw06_03_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
sw06_04_t2 ~ c("m4a", "m4b")*1 # This estimates the intercept of X4 at T2
'
lcs_swls_group_12_weak <- sem(mi_lcs_swls_group_weak, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
# strong invariance
mi_lcs_swls_group_strong <- '
swls_t1 =~ 1*sw06_01_t1 + c("lamb2", "lamb2")*sw06_02_t1 + c("lamb3", "lamb3")*sw06_03_t1 + c("lamb4", "lamb4")*sw06_04_t1 # This specifies the measurement model for swls_t1
swls_t2 =~ 1*sw06_01_t2 + c("lamb2", "lamb2")*sw06_02_t2 + c("lamb3", "lamb3")*sw06_03_t2 + c("lamb4", "lamb4")*sw06_04_t2 # This specifies the measurement model for swls_t2 with the equality constrained factor loadings
swls_t2 ~ 1*swls_t1 # This parameter regresses swls_t2 perfectly on swls_t1
d_swls_1 =~ 1*swls_t2 # This defines the latent change score factor as measured perfectly by scores on swls_t2
swls_t2 ~ 0*1 # This line constrains the intercept of swls_t2 to 0
swls_t2 ~~ 0*swls_t2 # This fixes the variance of swls_t2 to 0
d_swls_1 ~ c("d_int", "d_int")*1 # This estimates the intercept of the change score
swls_t1 ~ c("wb_int", "wb_int")*1 # This estimates the intercept of swls_t1
d_swls_1 ~~ c("d_var_a", "d_var_b")*d_swls_1 # This estimates the variance of the change scores
swls_t1 ~~ c("wb_var_a", "wb_var_b")*swls_t1 # This estimates the variance of the swls_t1
d_swls_1 ~~ c("fb_a", "fb_b")*swls_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
sw06_01_t1 ~~ c("cov1a", "cov1b")*sw06_01_t2 # This allows residual covariance on indicator X1 across T1 and T2
sw06_02_t1 ~~ c("cov2a", "cov2b")*sw06_02_t2 # This allows residual covariance on indicator X2 across T1 and T2
sw06_03_t1 ~~ c("cov3a", "cov3b")*sw06_03_t2 # This allows residual covariance on indicator X3 across T1 and T2
sw06_04_t1 ~~ c("cov4a", "cov4b")*sw06_04_t2 # This allows residual covariance on indicator X4 across T1 and T2
sw06_01_t1 ~~ c("res1a", "res1b")*sw06_01_t1 # This allows residual variance on indicator X1 at T1
sw06_02_t1 ~~ c("res2a", "res2b")*sw06_02_t1 # This allows residual variance on indicator X2 at T1
sw06_03_t1 ~~ c("res3a", "res3b")*sw06_03_t1 # This allows residual variance on indicator X3 at T1
sw06_04_t1 ~~ c("res4a", "res4b")*sw06_04_t1 # This allows residual variance on indicator X4 at T1
sw06_01_t2 ~~ c("res1a", "res1b")*sw06_01_t2 # This allows residual variance on indicator X1 at T2
sw06_02_t2 ~~ c("res2a", "res2b")*sw06_02_t2 # This allows residual variance on indicator X2 at T2
sw06_03_t2 ~~ c("res3a", "res3b")*sw06_03_t2 # This allows residual variance on indicator X3 at T2
sw06_04_t2 ~~ c("res4a", "res4b")*sw06_04_t2 # This allows residual variance on indicator X4 at T2
sw06_01_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
sw06_02_t1 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T1
sw06_03_t1 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T1
sw06_04_t1 ~ c("m4", "m4")*1 # This estimates the intercept of X4 at T1
sw06_01_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
sw06_02_t2 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T2
sw06_03_t2 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T2
sw06_04_t2 ~ c("m4", "m4")*1 # This estimates the intercept of X4 at T2
'
lcs_swls_group_12_strong <- sem(mi_lcs_swls_group_strong, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))Life satisfaction: results
# model comparison tests for measurement invariance
lavTestLRT(lcs_swls_group_12_config, lcs_swls_group_12_weak, lcs_swls_group_12_strong)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff Pr(>Chisq)
lcs_swls_group_12_config 50 6867.5 7014.6 94.812
lcs_swls_group_12_weak 53 6865.6 7001.0 98.864 4.2444 3 0.2363
lcs_swls_group_12_strong 58 6862.7 6978.8 106.007 7.1600 5 0.2090# show model with varying latent change parameters
# -> key parameter is "d_swls_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_lcs_swls_group_12_weak <- broom::tidy(lcs_swls_group_12_weak, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_swls_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_lcs_swls_group_12_weak, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_swls_1 ~1 | d_int_a | 0.187 | 0.317 | 3.237 | 0.001 | |
| d_swls_1 ~1 | d_int_b | 0.259 | 0.512 | 4.843 | 0.000 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(lcs_swls_group_12_strong, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_swls_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_swls_1 ~1 | d_int | 0.228 | 0.385 | 5.83 | 0 |
| d_swls_1 ~1 | d_int | 0.228 | 0.448 | 5.83 | 0 |
# whole model (weak invariance)
summary(lcs_swls_group_12_weak, fit.measures=TRUE, standardized=TRUE, rsquare=F)lavaan 0.6.15 ended normally after 55 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 58
Number of equality constraints 23
Number of observations per group:
Group 1 175
Group 2 179
Number of missing patterns per group:
Group 1 3
Group 2 2
Model Test User Model:
Standard Scaled
Test Statistic 98.864 91.108
Degrees of freedom 53 53
P-value (Chi-square) 0.000 0.001
Scaling correction factor 1.085
Yuan-Bentler correction (Mplus variant)
Test statistic for each group:
Group 1 43.450 40.041
Group 2 55.414 51.066
Model Test Baseline Model:
Test statistic 1778.340 1361.324
Degrees of freedom 56 56
P-value 0.000 0.000
Scaling correction factor 1.306
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.973 0.971
Tucker-Lewis Index (TLI) 0.972 0.969
Robust Comparative Fit Index (CFI) 0.976
Robust Tucker-Lewis Index (TLI) 0.975
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -3397.791 -3397.791
Scaling correction factor 0.671
for the MLR correction
Loglikelihood unrestricted model (H1) -3348.359 -3348.359
Scaling correction factor 1.096
for the MLR correction
Akaike (AIC) 6865.582 6865.582
Bayesian (BIC) 7001.007 7001.007
Sample-size adjusted Bayesian (SABIC) 6889.972 6889.972
Root Mean Square Error of Approximation:
RMSEA 0.070 0.064
90 Percent confidence interval - lower 0.048 0.042
90 Percent confidence interval - upper 0.091 0.085
P-value H_0: RMSEA <= 0.050 0.065 0.141
P-value H_0: RMSEA >= 0.080 0.229 0.104
Robust RMSEA 0.069
90 Percent confidence interval - lower 0.043
90 Percent confidence interval - upper 0.093
P-value H_0: Robust RMSEA <= 0.050 0.108
P-value H_0: Robust RMSEA >= 0.080 0.238
Standardized Root Mean Square Residual:
SRMR 0.047 0.047
Parameter Estimates:
Standard errors Sandwich
Information bread Observed
Observed information based on Hessian
Group 1 [Group 1]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t1 =~
s06_01_ 1.000 0.990 0.831
s06_02_ (lmb2) 0.957 0.040 24.191 0.000 0.947 0.793
s06_03_ (lmb3) 1.033 0.049 21.086 0.000 1.022 0.844
s06_04_ (lmb4) 0.855 0.054 15.877 0.000 0.846 0.649
swls_t2 =~
s06_01_ 1.000 0.953 0.821
s06_02_ (lmb2) 0.957 0.040 24.191 0.000 0.912 0.782
s06_03_ (lmb3) 1.033 0.049 21.086 0.000 0.985 0.835
s06_04_ (lmb4) 0.855 0.054 15.877 0.000 0.815 0.635
d_swls_1 =~
swls_t2 1.000 0.617 0.617
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t2 ~
swls_t1 1.000 1.038 1.038
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t1 ~~
d_swl_1 (fb_a) -0.209 0.055 -3.795 0.000 -0.358 -0.358
.sw06_01_t1 ~~
.s06_01_ (cov1) 0.176 0.064 2.739 0.006 0.176 0.402
.sw06_02_t1 ~~
.s06_02_ (cov2) 0.228 0.066 3.438 0.001 0.228 0.431
.sw06_03_t1 ~~
.s06_03_ (cov3) 0.113 0.059 1.899 0.058 0.113 0.267
.sw06_04_t1 ~~
.s06_04_ (cov4) 0.406 0.088 4.611 0.000 0.406 0.414
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.swls_t2 0.000 0.000 0.000
d_swl_1 (d_n_) 0.187 0.058 3.237 0.001 0.317 0.317
swls_t1 (wb__) 2.801 0.089 31.293 0.000 2.830 2.830
.s06_01_ 0.000 0.000 0.000
.s06_02_ (m2a) 0.151 0.130 1.164 0.245 0.151 0.126
.s06_03_ (m3a) 0.021 0.156 0.133 0.894 0.021 0.017
.s06_04_ (m4a) 0.525 0.166 3.164 0.002 0.525 0.403
.s06_01_ 0.000 0.000 0.000
.s06_02_ (m2a) 0.151 0.130 1.164 0.245 0.151 0.129
.s06_03_ (m3a) 0.021 0.156 0.133 0.894 0.021 0.018
.s06_04_ (m4a) 0.525 0.166 3.164 0.002 0.525 0.409
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.swls_t2 0.000 0.000 0.000
d_swl_1 (d_v_) 0.346 0.087 3.960 0.000 1.000 1.000
swls_t1 (wb__) 0.979 0.104 9.396 0.000 1.000 1.000
.s06_01_ (res1) 0.439 0.063 7.015 0.000 0.439 0.309
.s06_02_ (res2) 0.530 0.063 8.458 0.000 0.530 0.371
.s06_03_ (res3) 0.421 0.067 6.306 0.000 0.421 0.287
.s06_04_ (res4) 0.982 0.100 9.841 0.000 0.982 0.578
.s06_01_ (res1) 0.439 0.063 7.015 0.000 0.439 0.326
.s06_02_ (res2) 0.530 0.063 8.458 0.000 0.530 0.389
.s06_03_ (res3) 0.421 0.067 6.306 0.000 0.421 0.303
.s06_04_ (res4) 0.982 0.100 9.841 0.000 0.982 0.596
Group 2 [Group 2]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t1 =~
s06_01_ 1.000 0.993 0.827
s06_02_ (lmb2) 0.957 0.040 24.191 0.000 0.951 0.800
s06_03_ (lmb3) 1.033 0.049 21.086 0.000 1.026 0.857
s06_04_ (lmb4) 0.855 0.054 15.877 0.000 0.850 0.703
swls_t2 =~
s06_01_ 1.000 1.001 0.829
s06_02_ (lmb2) 0.957 0.040 24.191 0.000 0.958 0.802
s06_03_ (lmb3) 1.033 0.049 21.086 0.000 1.034 0.858
s06_04_ (lmb4) 0.855 0.054 15.877 0.000 0.856 0.706
d_swls_1 =~
swls_t2 1.000 0.506 0.506
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t2 ~
swls_t1 1.000 0.992 0.992
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t1 ~~
d_swl_1 (fb_b) -0.121 0.053 -2.292 0.022 -0.240 -0.240
.sw06_01_t1 ~~
.s06_01_ (cv1b) 0.176 0.058 3.059 0.002 0.176 0.386
.sw06_02_t1 ~~
.s06_02_ (cv2b) 0.195 0.061 3.210 0.001 0.195 0.383
.sw06_03_t1 ~~
.s06_03_ (cv3b) 0.030 0.055 0.539 0.590 0.030 0.077
.sw06_04_t1 ~~
.s06_04_ (cv4b) 0.207 0.076 2.724 0.006 0.207 0.280
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.swls_t2 0.000 0.000 0.000
d_swl_1 (d_n_) 0.259 0.053 4.843 0.000 0.512 0.512
swls_t1 (wb__) 2.544 0.087 29.309 0.000 2.561 2.561
.s06_01_ 0.000 0.000 0.000
.s06_02_ (m2b) 0.274 0.122 2.251 0.024 0.274 0.231
.s06_03_ (m3b) 0.198 0.143 1.387 0.165 0.198 0.166
.s06_04_ (m4b) 0.573 0.158 3.632 0.000 0.573 0.475
.s06_01_ 0.000 0.000 0.000
.s06_02_ (m2b) 0.274 0.122 2.251 0.024 0.274 0.230
.s06_03_ (m3b) 0.198 0.143 1.387 0.165 0.198 0.165
.s06_04_ (m4b) 0.573 0.158 3.632 0.000 0.573 0.473
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.swls_t2 0.000 0.000 0.000
d_swl_1 (d_v_) 0.257 0.057 4.467 0.000 1.000 1.000
swls_t1 (wb__) 0.987 0.105 9.396 0.000 1.000 1.000
.s06_01_ (rs1b) 0.456 0.063 7.279 0.000 0.456 0.316
.s06_02_ (rs2b) 0.508 0.061 8.280 0.000 0.508 0.360
.s06_03_ (rs3b) 0.382 0.056 6.784 0.000 0.382 0.266
.s06_04_ (rs4b) 0.737 0.080 9.220 0.000 0.737 0.505
.s06_01_ (rs1b) 0.456 0.063 7.279 0.000 0.456 0.313
.s06_02_ (rs2b) 0.508 0.061 8.280 0.000 0.508 0.356
.s06_03_ (rs3b) 0.382 0.056 6.784 0.000 0.382 0.263
.s06_04_ (rs4b) 0.737 0.080 9.220 0.000 0.737 0.501Meaning in life: fitting multi-group models
# adapt latent change score model from above and add grouping factor in estimation (also add vectorized equality constraints to the model)
# configural invariance
mi_lcs_meaning_group_config <- '
meaning_t1 =~ 1*meaning_par1_t1 + c("lamb2a", "lamb2b")*meaning_par2_t1 + c("lamb3a", "lamb3b")*meaning_par3_t1 # This specifies the measurement model for meaning_t1
meaning_t2 =~ 1*meaning_par1_t2 + c("lamb2a", "lamb2b")*meaning_par2_t2 + c("lamb3a", "lamb3b")*meaning_par3_t2 # This specifies the measurement model for meaning_t2 with the equality constrained factor loadings
meaning_t2 ~ 1*meaning_t1 # This parameter regresses meaning_t2 perfectly on meaning_t1
d_meaning_1 =~ 1*meaning_t2 # This defines the latent change score factor as measured perfectly by scores on meaning_t2
meaning_t2 ~ 0*1 # This line constrains the intercept of meaning_t2 to 0
meaning_t2 ~~ 0*meaning_t2 # This fixes the variance of meaning_t2 to 0
d_meaning_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
meaning_t1 ~ c("wb_int_a", "wb_int_b")*1 # This estimates the intercept of meaning_t1
d_meaning_1 ~~ c("d_var_a", "d_var_b")*d_meaning_1 # This estimates the variance of the change scores
meaning_t1 ~~ c("wb_var_a", "wb_var_b")*meaning_t1 # This estimates the variance of the meaning_t1
d_meaning_1 ~~ c("fb_a", "fb_b")*meaning_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
meaning_par1_t1 ~~ c("cov1a", "cov1b")*meaning_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
meaning_par2_t1 ~~ c("cov2a", "cov2b")*meaning_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
meaning_par3_t1 ~~ c("cov3a", "cov3b")*meaning_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
meaning_par1_t1 ~~ c("res1a", "res1b")*meaning_par1_t1 # This allows residual variance on indicator X1 at T1
meaning_par2_t1 ~~ c("res2a", "res2b")*meaning_par2_t1 # This allows residual variance on indicator X2 at T1
meaning_par3_t1 ~~ c("res3a", "res3b")*meaning_par3_t1 # This allows residual variance on indicator X3 at T1
meaning_par1_t2 ~~ c("res1a", "res1b")*meaning_par1_t2 # This allows residual variance on indicator X1 at T2
meaning_par2_t2 ~~ c("res2a", "res2b")*meaning_par2_t2 # This allows residual variance on indicator X2 at T2
meaning_par3_t2 ~~ c("res3a", "res3b")*meaning_par3_t2 # This allows residual variance on indicator X3 at T2
meaning_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
meaning_par2_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
meaning_par3_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
meaning_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
meaning_par2_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
meaning_par3_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
'
lcs_meaning_group_12_config <- sem(mi_lcs_meaning_group_config, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr',
fixed.x=FALSE, missing='fiml', group = "group")
# weak invariance
mi_lcs_meaning_group_weak <- '
meaning_t1 =~ 1*meaning_par1_t1 + c("lamb2", "lamb2")*meaning_par2_t1 + c("lamb3", "lamb3")*meaning_par3_t1 # This specifies the measurement model for meaning_t1
meaning_t2 =~ 1*meaning_par1_t2 + c("lamb2", "lamb2")*meaning_par2_t2 + c("lamb3", "lamb3")*meaning_par3_t2 # This specifies the measurement model for meaning_t2 with the equality constrained factor loadings
meaning_t2 ~ 1*meaning_t1 # This parameter regresses meaning_t2 perfectly on meaning_t1
d_meaning_1 =~ 1*meaning_t2 # This defines the latent change score factor as measured perfectly by scores on meaning_t2
meaning_t2 ~ 0*1 # This line constrains the intercept of meaning_t2 to 0
meaning_t2 ~~ 0*meaning_t2 # This fixes the variance of meaning_t2 to 0
d_meaning_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
meaning_t1 ~ c("wb_int_a", "wb_int_b")*1 # This estimates the intercept of meaning_t1
d_meaning_1 ~~ c("d_var_a", "d_var_b")*d_meaning_1 # This estimates the variance of the change scores
meaning_t1 ~~ c("wb_var_a", "wb_var_b")*meaning_t1 # This estimates the variance of the meaning_t1
d_meaning_1 ~~ c("fb_a", "fb_b")*meaning_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
meaning_par1_t1 ~~ c("cov1a", "cov1b")*meaning_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
meaning_par2_t1 ~~ c("cov2a", "cov2b")*meaning_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
meaning_par3_t1 ~~ c("cov3a", "cov3b")*meaning_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
meaning_par1_t1 ~~ c("res1a", "res1b")*meaning_par1_t1 # This allows residual variance on indicator X1 at T1
meaning_par2_t1 ~~ c("res2a", "res2b")*meaning_par2_t1 # This allows residual variance on indicator X2 at T1
meaning_par3_t1 ~~ c("res3a", "res3b")*meaning_par3_t1 # This allows residual variance on indicator X3 at T1
meaning_par1_t2 ~~ c("res1a", "res1b")*meaning_par1_t2 # This allows residual variance on indicator X1 at T2
meaning_par2_t2 ~~ c("res2a", "res2b")*meaning_par2_t2 # This allows residual variance on indicator X2 at T2
meaning_par3_t2 ~~ c("res3a", "res3b")*meaning_par3_t2 # This allows residual variance on indicator X3 at T2
meaning_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
meaning_par2_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
meaning_par3_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
meaning_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
meaning_par2_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
meaning_par3_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
'
lcs_meaning_group_12_weak <- sem(mi_lcs_meaning_group_weak, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
# strong invariance
mi_lcs_meaning_group_strong <- '
meaning_t1 =~ 1*meaning_par1_t1 + c("lamb2", "lamb2")*meaning_par2_t1 + c("lamb3", "lamb3")*meaning_par3_t1 # This specifies the measurement model for meaning_t1
meaning_t2 =~ 1*meaning_par1_t2 + c("lamb2", "lamb2")*meaning_par2_t2 + c("lamb3", "lamb3")*meaning_par3_t2 # This specifies the measurement model for meaning_t2 with the equality constrained factor loadings
meaning_t2 ~ 1*meaning_t1 # This parameter regresses meaning_t2 perfectly on meaning_t1
d_meaning_1 =~ 1*meaning_t2 # This defines the latent change score factor as measured perfectly by scores on meaning_t2
meaning_t2 ~ 0*1 # This line constrains the intercept of meaning_t2 to 0
meaning_t2 ~~ 0*meaning_t2 # This fixes the variance of meaning_t2 to 0
d_meaning_1 ~ c("d_int", "d_int")*1 # This estimates the intercept of the change score
meaning_t1 ~ c("wb_int", "wb_int")*1 # This estimates the intercept of meaning_t1
d_meaning_1 ~~ c("d_var_a", "d_var_b")*d_meaning_1 # This estimates the variance of the change scores
meaning_t1 ~~ c("wb_var_a", "wb_var_b")*meaning_t1 # This estimates the variance of the meaning_t1
d_meaning_1 ~~ c("fb_a", "fb_b")*meaning_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
meaning_par1_t1 ~~ c("cov1a", "cov1b")*meaning_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
meaning_par2_t1 ~~ c("cov2a", "cov2b")*meaning_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
meaning_par3_t1 ~~ c("cov3a", "cov3b")*meaning_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
meaning_par1_t1 ~~ c("res1a", "res1b")*meaning_par1_t1 # This allows residual variance on indicator X1 at T1
meaning_par2_t1 ~~ c("res2a", "res2b")*meaning_par2_t1 # This allows residual variance on indicator X2 at T1
meaning_par3_t1 ~~ c("res3a", "res3b")*meaning_par3_t1 # This allows residual variance on indicator X3 at T1
meaning_par1_t2 ~~ c("res1a", "res1b")*meaning_par1_t2 # This allows residual variance on indicator X1 at T2
meaning_par2_t2 ~~ c("res2a", "res2b")*meaning_par2_t2 # This allows residual variance on indicator X2 at T2
meaning_par3_t2 ~~ c("res3a", "res3b")*meaning_par3_t2 # This allows residual variance on indicator X3 at T2
meaning_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
meaning_par2_t1 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T1
meaning_par3_t1 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T1
meaning_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
meaning_par2_t2 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T2
meaning_par3_t2 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T2
'
lcs_meaning_group_12_strong <- sem(mi_lcs_meaning_group_strong, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))Meaning in life: results
# model comparison tests for measurement invariance
lavTestLRT(lcs_meaning_group_12_config, lcs_meaning_group_12_weak, lcs_meaning_group_12_strong)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
lcs_meaning_group_12_config 24 5087.3 5203.4 22.780
lcs_meaning_group_12_weak 26 5087.8 5196.2 27.304 4.7694 2
lcs_meaning_group_12_strong 30 5084.2 5177.1 31.705 4.3666 4
Pr(>Chisq)
lcs_meaning_group_12_config
lcs_meaning_group_12_weak 0.09212 .
lcs_meaning_group_12_strong 0.35867
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_meaning_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_lcs_meaning_group_12_weak <- broom::tidy(lcs_meaning_group_12_weak, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_meaning_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_lcs_meaning_group_12_weak, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_meaning_1 ~1 | d_int_a | 0.105 | 0.186 | 1.961 | 0.050 | |
| d_meaning_1 ~1 | d_int_b | 0.113 | 0.200 | 2.031 | 0.042 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(lcs_meaning_group_12_strong, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_meaning_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_meaning_1 ~1 | d_int | 0.109 | 0.193 | 2.776 | 0.005 |
| d_meaning_1 ~1 | d_int | 0.109 | 0.194 | 2.776 | 0.005 |
# whole model (weak invariance)
summary(lcs_meaning_group_12_weak, fit.measures=TRUE, standardized=TRUE, rsquare=F)lavaan 0.6.15 ended normally after 59 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 44
Number of equality constraints 16
Number of observations per group:
Group 1 175
Group 2 179
Number of missing patterns per group:
Group 1 3
Group 2 2
Model Test User Model:
Standard Scaled
Test Statistic 27.304 26.209
Degrees of freedom 26 26
P-value (Chi-square) 0.394 0.452
Scaling correction factor 1.042
Yuan-Bentler correction (Mplus variant)
Test statistic for each group:
Group 1 10.565 10.141
Group 2 16.739 16.067
Model Test Baseline Model:
Test statistic 1490.873 1210.255
Degrees of freedom 30 30
P-value 0.000 0.000
Scaling correction factor 1.232
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.999 1.000
Tucker-Lewis Index (TLI) 0.999 1.000
Robust Comparative Fit Index (CFI) 1.000
Robust Tucker-Lewis Index (TLI) 1.000
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -2515.917 -2515.917
Scaling correction factor 0.749
for the MLR correction
Loglikelihood unrestricted model (H1) -2502.265 -2502.265
Scaling correction factor 1.112
for the MLR correction
Akaike (AIC) 5087.834 5087.834
Bayesian (BIC) 5196.174 5196.174
Sample-size adjusted Bayesian (SABIC) 5107.347 5107.347
Root Mean Square Error of Approximation:
RMSEA 0.017 0.007
90 Percent confidence interval - lower 0.000 0.000
90 Percent confidence interval - upper 0.063 0.059
P-value H_0: RMSEA <= 0.050 0.853 0.888
P-value H_0: RMSEA >= 0.080 0.005 0.003
Robust RMSEA 0.000
90 Percent confidence interval - lower 0.000
90 Percent confidence interval - upper 0.063
P-value H_0: Robust RMSEA <= 0.050 0.861
P-value H_0: Robust RMSEA >= 0.080 0.008
Standardized Root Mean Square Residual:
SRMR 0.037 0.037
Parameter Estimates:
Standard errors Sandwich
Information bread Observed
Observed information based on Hessian
Group 1 [Group 1]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t1 =~
mnn_1_1 1.000 0.976 0.755
mnn_2_1 (lmb2) 1.091 0.052 20.774 0.000 1.064 0.879
mnn_3_1 (lmb3) 1.219 0.063 19.290 0.000 1.190 0.916
meaning_t2 =~
mnn_1_2 1.000 0.967 0.751
mnn_2_2 (lmb2) 1.091 0.052 20.774 0.000 1.054 0.877
mnn_3_2 (lmb3) 1.219 0.063 19.290 0.000 1.179 0.914
d_meaning_1 =~
mnng_t2 1.000 0.586 0.586
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t2 ~
meaning_t1 1.000 1.010 1.010
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t1 ~~
d_mnn_1 (fb_a) -0.170 0.062 -2.732 0.006 -0.307 -0.307
.meaning_par1_t1 ~~
.mnn_1_2 (cov1) 0.458 0.068 6.748 0.000 0.458 0.635
.meaning_par2_t1 ~~
.mnn_2_2 (cov2) 0.036 0.045 0.804 0.421 0.036 0.108
.meaning_par3_t1 ~~
.mnn_3_2 (cov3) 0.047 0.063 0.743 0.457 0.047 0.172
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.mnng_t2 0.000 0.000 0.000
d_mnn_1 (d_n_) 0.105 0.054 1.961 0.050 0.186 0.186
mnng_t1 (wb__) 3.447 0.091 37.687 0.000 3.532 3.532
.mnn_1_1 0.000 0.000 0.000
.mnn_2_1 (m2a) 0.175 0.198 0.883 0.377 0.175 0.144
.mnn_3_1 (m3a) -0.182 0.229 -0.794 0.427 -0.182 -0.140
.mnn_1_2 0.000 0.000 0.000
.mnn_2_2 (m2a) 0.175 0.198 0.883 0.377 0.175 0.146
.mnn_3_2 (m3a) -0.182 0.229 -0.794 0.427 -0.182 -0.141
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.mnng_t2 0.000 0.000 0.000
d_mnn_1 (d_v_) 0.321 0.061 5.276 0.000 1.000 1.000
mnng_t1 (wb__) 0.953 0.145 6.554 0.000 1.000 1.000
.mnn_1_1 (res1) 0.721 0.070 10.269 0.000 0.721 0.431
.mnn_2_1 (res2) 0.334 0.052 6.368 0.000 0.334 0.228
.mnn_3_1 (res3) 0.273 0.080 3.395 0.001 0.273 0.162
.mnn_1_2 (res1) 0.721 0.070 10.269 0.000 0.721 0.435
.mnn_2_2 (res2) 0.334 0.052 6.368 0.000 0.334 0.231
.mnn_3_2 (res3) 0.273 0.080 3.395 0.001 0.273 0.164
Group 2 [Group 2]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t1 =~
mnn_1_1 1.000 0.932 0.738
mnn_2_1 (lmb2) 1.091 0.052 20.774 0.000 1.016 0.857
mnn_3_1 (lmb3) 1.219 0.063 19.290 0.000 1.137 0.888
meaning_t2 =~
mnn_1_2 1.000 0.913 0.730
mnn_2_2 (lmb2) 1.091 0.052 20.774 0.000 0.995 0.853
mnn_3_2 (lmb3) 1.219 0.063 19.290 0.000 1.113 0.884
d_meaning_1 =~
mnng_t2 1.000 0.618 0.618
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t2 ~
meaning_t1 1.000 1.021 1.021
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t1 ~~
d_mnn_1 (fb_b) -0.177 0.077 -2.293 0.022 -0.336 -0.336
.meaning_par1_t1 ~~
.mnn_1_2 (cv1b) 0.417 0.089 4.706 0.000 0.417 0.573
.meaning_par2_t1 ~~
.mnn_2_2 (cv2b) 0.104 0.049 2.119 0.034 0.104 0.280
.meaning_par3_t1 ~~
.mnn_3_2 (cv3b) 0.068 0.056 1.210 0.226 0.068 0.197
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.mnng_t2 0.000 0.000 0.000
d_mnn_1 (d_n_) 0.113 0.056 2.031 0.042 0.200 0.200
mnng_t1 (wb__) 3.353 0.090 37.059 0.000 3.597 3.597
.mnn_1_1 0.000 0.000 0.000
.mnn_2_1 (m2b) 0.057 0.191 0.301 0.763 0.057 0.048
.mnn_3_1 (m3b) -0.292 0.228 -1.283 0.200 -0.292 -0.228
.mnn_1_2 0.000 0.000 0.000
.mnn_2_2 (m2b) 0.057 0.191 0.301 0.763 0.057 0.049
.mnn_3_2 (m3b) -0.292 0.228 -1.283 0.200 -0.292 -0.232
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.mnng_t2 0.000 0.000 0.000
d_mnn_1 (d_v_) 0.318 0.085 3.730 0.000 1.000 1.000
mnng_t1 (wb__) 0.869 0.134 6.484 0.000 1.000 1.000
.mnn_1_1 (rs1b) 0.728 0.087 8.416 0.000 0.728 0.456
.mnn_2_1 (rs2b) 0.372 0.055 6.759 0.000 0.372 0.265
.mnn_3_1 (rs3b) 0.345 0.061 5.688 0.000 0.345 0.211
.mnn_1_2 (rs1b) 0.728 0.087 8.416 0.000 0.728 0.466
.mnn_2_2 (rs2b) 0.372 0.055 6.759 0.000 0.372 0.273
.mnn_3_2 (rs3b) 0.345 0.061 5.688 0.000 0.345 0.218Self-esteem: fitting multi-group models
# adapt latent change score model from above and add grouping factor in estimation (also add vectorized equality constraints to the model)
# configural invariance
mi_lcs_selfes_group_config <- '
selfes_t1 =~ 1*selfes_par1_t1 + c("lamb2a", "lamb2b")*selfes_par2_t1 + c("lamb3a", "lamb3b")*selfes_par3_t1 # This specifies the measurement model for selfes_t1
selfes_t2 =~ 1*selfes_par1_t2 + c("lamb2a", "lamb2b")*selfes_par2_t2 + c("lamb3a", "lamb3b")*selfes_par3_t2 # This specifies the measurement model for selfes_t2 with the equality constrained factor loadings
selfes_t2 ~ 1*selfes_t1 # This parameter regresses selfes_t2 perfectly on selfes_t1
d_selfes_1 =~ 1*selfes_t2 # This defines the latent change score factor as measured perfectly by scores on selfes_t2
selfes_t2 ~ 0*1 # This line constrains the intercept of selfes_t2 to 0
selfes_t2 ~~ 0*selfes_t2 # This fixes the variance of selfes_t2 to 0
d_selfes_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
selfes_t1 ~ c("wb_int_a", "wb_int_b")*1 # This estimates the intercept of selfes_t1
d_selfes_1 ~~ c("d_var_a", "d_var_b")*d_selfes_1 # This estimates the variance of the change scores
selfes_t1 ~~ c("wb_var_a", "wb_var_b")*selfes_t1 # This estimates the variance of the selfes_t1
d_selfes_1 ~~ c("fb_a", "fb_b")*selfes_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
selfes_par1_t1 ~~ c("cov1a", "cov1b")*selfes_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
selfes_par2_t1 ~~ c("cov2a", "cov2b")*selfes_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
selfes_par3_t1 ~~ c("cov3a", "cov3b")*selfes_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
selfes_par1_t1 ~~ c("res1a", "res1b")*selfes_par1_t1 # This allows residual variance on indicator X1 at T1
selfes_par2_t1 ~~ c("res2a", "res2b")*selfes_par2_t1 # This allows residual variance on indicator X2 at T1
selfes_par3_t1 ~~ c("res3a", "res3b")*selfes_par3_t1 # This allows residual variance on indicator X3 at T1
selfes_par1_t2 ~~ c("res1a", "res1b")*selfes_par1_t2 # This allows residual variance on indicator X1 at T2
selfes_par2_t2 ~~ c("res2a", "res2b")*selfes_par2_t2 # This allows residual variance on indicator X2 at T2
selfes_par3_t2 ~~ c("res3a", "res3b")*selfes_par3_t2 # This allows residual variance on indicator X3 at T2
selfes_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
selfes_par2_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
selfes_par3_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
selfes_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
selfes_par2_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
selfes_par3_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
'
lcs_selfes_group_12_config <- sem(mi_lcs_selfes_group_config, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr',
fixed.x=FALSE, missing='fiml', group = "group")
# weak invariance
mi_lcs_selfes_group_weak <- '
selfes_t1 =~ 1*selfes_par1_t1 + c("lamb2", "lamb2")*selfes_par2_t1 + c("lamb3", "lamb3")*selfes_par3_t1 # This specifies the measurement model for selfes_t1
selfes_t2 =~ 1*selfes_par1_t2 + c("lamb2", "lamb2")*selfes_par2_t2 + c("lamb3", "lamb3")*selfes_par3_t2 # This specifies the measurement model for selfes_t2 with the equality constrained factor loadings
selfes_t2 ~ 1*selfes_t1 # This parameter regresses selfes_t2 perfectly on selfes_t1
d_selfes_1 =~ 1*selfes_t2 # This defines the latent change score factor as measured perfectly by scores on selfes_t2
selfes_t2 ~ 0*1 # This line constrains the intercept of selfes_t2 to 0
selfes_t2 ~~ 0*selfes_t2 # This fixes the variance of selfes_t2 to 0
d_selfes_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
selfes_t1 ~ c("wb_int_a", "wb_int_b")*1 # This estimates the intercept of selfes_t1
d_selfes_1 ~~ c("d_var_a", "d_var_b")*d_selfes_1 # This estimates the variance of the change scores
selfes_t1 ~~ c("wb_var_a", "wb_var_b")*selfes_t1 # This estimates the variance of the selfes_t1
d_selfes_1 ~~ c("fb_a", "fb_b")*selfes_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
selfes_par1_t1 ~~ c("cov1a", "cov1b")*selfes_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
selfes_par2_t1 ~~ c("cov2a", "cov2b")*selfes_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
selfes_par3_t1 ~~ c("cov3a", "cov3b")*selfes_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
selfes_par1_t1 ~~ c("res1a", "res1b")*selfes_par1_t1 # This allows residual variance on indicator X1 at T1
selfes_par2_t1 ~~ c("res2a", "res2b")*selfes_par2_t1 # This allows residual variance on indicator X2 at T1
selfes_par3_t1 ~~ c("res3a", "res3b")*selfes_par3_t1 # This allows residual variance on indicator X3 at T1
selfes_par1_t2 ~~ c("res1a", "res1b")*selfes_par1_t2 # This allows residual variance on indicator X1 at T2
selfes_par2_t2 ~~ c("res2a", "res2b")*selfes_par2_t2 # This allows residual variance on indicator X2 at T2
selfes_par3_t2 ~~ c("res3a", "res3b")*selfes_par3_t2 # This allows residual variance on indicator X3 at T2
selfes_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
selfes_par2_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
selfes_par3_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
selfes_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
selfes_par2_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
selfes_par3_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
'
lcs_selfes_group_12_weak <- sem(mi_lcs_selfes_group_weak, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
# strong invariance
mi_lcs_selfes_group_strong <- '
selfes_t1 =~ 1*selfes_par1_t1 + c("lamb2", "lamb2")*selfes_par2_t1 + c("lamb3", "lamb3")*selfes_par3_t1 # This specifies the measurement model for selfes_t1
selfes_t2 =~ 1*selfes_par1_t2 + c("lamb2", "lamb2")*selfes_par2_t2 + c("lamb3", "lamb3")*selfes_par3_t2 # This specifies the measurement model for selfes_t2 with the equality constrained factor loadings
selfes_t2 ~ 1*selfes_t1 # This parameter regresses selfes_t2 perfectly on selfes_t1
d_selfes_1 =~ 1*selfes_t2 # This defines the latent change score factor as measured perfectly by scores on selfes_t2
selfes_t2 ~ 0*1 # This line constrains the intercept of selfes_t2 to 0
selfes_t2 ~~ 0*selfes_t2 # This fixes the variance of selfes_t2 to 0
d_selfes_1 ~ c("d_int", "d_int")*1 # This estimates the intercept of the change score
selfes_t1 ~ c("wb_int", "wb_int")*1 # This estimates the intercept of selfes_t1
d_selfes_1 ~~ c("d_var_a", "d_var_b")*d_selfes_1 # This estimates the variance of the change scores
selfes_t1 ~~ c("wb_var_a", "wb_var_b")*selfes_t1 # This estimates the variance of the selfes_t1
d_selfes_1 ~~ c("fb_a", "fb_b")*selfes_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
selfes_par1_t1 ~~ c("cov1a", "cov1b")*selfes_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
selfes_par2_t1 ~~ c("cov2a", "cov2b")*selfes_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
selfes_par3_t1 ~~ c("cov3a", "cov3b")*selfes_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
selfes_par1_t1 ~~ c("res1a", "res1b")*selfes_par1_t1 # This allows residual variance on indicator X1 at T1
selfes_par2_t1 ~~ c("res2a", "res2b")*selfes_par2_t1 # This allows residual variance on indicator X2 at T1
selfes_par3_t1 ~~ c("res3a", "res3b")*selfes_par3_t1 # This allows residual variance on indicator X3 at T1
selfes_par1_t2 ~~ c("res1a", "res1b")*selfes_par1_t2 # This allows residual variance on indicator X1 at T2
selfes_par2_t2 ~~ c("res2a", "res2b")*selfes_par2_t2 # This allows residual variance on indicator X2 at T2
selfes_par3_t2 ~~ c("res3a", "res3b")*selfes_par3_t2 # This allows residual variance on indicator X3 at T2
selfes_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
selfes_par2_t1 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T1
selfes_par3_t1 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T1
selfes_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
selfes_par2_t2 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T2
selfes_par3_t2 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T2
'
lcs_selfes_group_12_strong <- sem(mi_lcs_selfes_group_strong, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))Meaning in life: results
# model comparison tests for measurement invariance
lavTestLRT(lcs_selfes_group_12_config, lcs_selfes_group_12_weak, lcs_selfes_group_12_strong)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
lcs_selfes_group_12_config 24 3696.0 3812.1 19.460
lcs_selfes_group_12_weak 26 3692.2 3800.5 19.616 0.1506 2
lcs_selfes_group_12_strong 30 3701.5 3794.3 36.898 16.7031 4
Pr(>Chisq)
lcs_selfes_group_12_config
lcs_selfes_group_12_weak 0.927464
lcs_selfes_group_12_strong 0.002207 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_selfes_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_lcs_selfes_group_12_weak <- broom::tidy(lcs_selfes_group_12_weak, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_selfes_1 ~1 ")) %>% mutate(sig_diff = "**")
kable(params_lcs_selfes_group_12_weak, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_selfes_1 ~1 | d_int_a | 0.108 | 0.247 | 2.765 | 0.006 | ** |
| d_selfes_1 ~1 | d_int_b | 0.166 | 0.398 | 4.158 | 0.000 | ** |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(lcs_selfes_group_12_strong, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_selfes_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_selfes_1 ~1 | d_int | 0.139 | 0.315 | 4.742 | 0 |
| d_selfes_1 ~1 | d_int | 0.139 | 0.331 | 4.742 | 0 |
# whole model (weak invariance)
summary(lcs_selfes_group_12_weak, fit.measures=TRUE, standardized=TRUE, rsquare=F)lavaan 0.6.15 ended normally after 66 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 44
Number of equality constraints 16
Number of observations per group:
Group 1 175
Group 2 179
Number of missing patterns per group:
Group 1 3
Group 2 2
Model Test User Model:
Standard Scaled
Test Statistic 19.616 18.369
Degrees of freedom 26 26
P-value (Chi-square) 0.809 0.862
Scaling correction factor 1.068
Yuan-Bentler correction (Mplus variant)
Test statistic for each group:
Group 1 4.565 4.274
Group 2 15.052 14.095
Model Test Baseline Model:
Test statistic 1755.372 1429.002
Degrees of freedom 30 30
P-value 0.000 0.000
Scaling correction factor 1.228
User Model versus Baseline Model:
Comparative Fit Index (CFI) 1.000 1.000
Tucker-Lewis Index (TLI) 1.004 1.006
Robust Comparative Fit Index (CFI) 1.000
Robust Tucker-Lewis Index (TLI) 1.005
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -1818.093 -1818.093
Scaling correction factor 0.681
for the MLR correction
Loglikelihood unrestricted model (H1) -1808.285 -1808.285
Scaling correction factor 1.069
for the MLR correction
Akaike (AIC) 3692.186 3692.186
Bayesian (BIC) 3800.527 3800.527
Sample-size adjusted Bayesian (SABIC) 3711.699 3711.699
Root Mean Square Error of Approximation:
RMSEA 0.000 0.000
90 Percent confidence interval - lower 0.000 0.000
90 Percent confidence interval - upper 0.038 0.031
P-value H_0: RMSEA <= 0.050 0.982 0.991
P-value H_0: RMSEA >= 0.080 0.000 0.000
Robust RMSEA 0.000
90 Percent confidence interval - lower 0.000
90 Percent confidence interval - upper 0.039
P-value H_0: Robust RMSEA <= 0.050 0.979
P-value H_0: Robust RMSEA >= 0.080 0.000
Standardized Root Mean Square Residual:
SRMR 0.031 0.031
Parameter Estimates:
Standard errors Sandwich
Information bread Observed
Observed information based on Hessian
Group 1 [Group 1]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t1 =~
slf_1_1 1.000 0.772 0.900
slf_2_1 (lmb2) 1.155 0.040 28.592 0.000 0.891 0.931
slf_3_1 (lmb3) 1.083 0.050 21.584 0.000 0.836 0.805
selfes_t2 =~
slf_1_2 1.000 0.760 0.897
slf_2_2 (lmb2) 1.155 0.040 28.592 0.000 0.877 0.929
slf_3_2 (lmb3) 1.083 0.050 21.584 0.000 0.823 0.801
d_selfes_1 =~
slfs_t2 1.000 0.577 0.577
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t2 ~
selfes_t1 1.000 1.016 1.016
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t1 ~~
d_slf_1 (fb_a) -0.105 0.027 -3.846 0.000 -0.311 -0.311
.selfes_par1_t1 ~~
.slf_1_2 (cov1) 0.072 0.019 3.783 0.000 0.072 0.516
.selfes_par2_t1 ~~
.slf_2_2 (cov2) 0.067 0.025 2.627 0.009 0.067 0.546
.selfes_par3_t1 ~~
.slf_3_2 (cov3) 0.130 0.037 3.463 0.001 0.130 0.342
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.slfs_t2 0.000 0.000 0.000
d_slf_1 (d_n_) 0.108 0.039 2.765 0.006 0.247 0.247
slfs_t1 (wb__) 3.674 0.064 57.373 0.000 4.759 4.759
.slf_1_1 0.000 0.000 0.000
.slf_2_1 (m2a) -0.838 0.164 -5.108 0.000 -0.838 -0.875
.slf_3_1 (m3a) -1.202 0.192 -6.253 0.000 -1.202 -1.158
.slf_1_2 0.000 0.000 0.000
.slf_2_2 (m2a) -0.838 0.164 -5.108 0.000 -0.838 -0.887
.slf_3_2 (m3a) -1.202 0.192 -6.253 0.000 -1.202 -1.169
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.slfs_t2 0.000 0.000 0.000
d_slf_1 (d_v_) 0.192 0.038 5.058 0.000 1.000 1.000
slfs_t1 (wb__) 0.596 0.063 9.423 0.000 1.000 1.000
.slf_1_1 (res1) 0.140 0.020 7.066 0.000 0.140 0.191
.slf_2_1 (res2) 0.122 0.025 4.849 0.000 0.122 0.133
.slf_3_1 (res3) 0.379 0.037 10.352 0.000 0.379 0.352
.slf_1_2 (res1) 0.140 0.020 7.066 0.000 0.140 0.196
.slf_2_2 (res2) 0.122 0.025 4.849 0.000 0.122 0.137
.slf_3_2 (res3) 0.379 0.037 10.352 0.000 0.379 0.359
Group 2 [Group 2]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t1 =~
slf_1_1 1.000 0.738 0.879
slf_2_1 (lmb2) 1.155 0.040 28.592 0.000 0.852 0.913
slf_3_1 (lmb3) 1.083 0.050 21.584 0.000 0.799 0.780
selfes_t2 =~
slf_1_2 1.000 0.751 0.883
slf_2_2 (lmb2) 1.155 0.040 28.592 0.000 0.867 0.916
slf_3_2 (lmb3) 1.083 0.050 21.584 0.000 0.814 0.786
d_selfes_1 =~
slfs_t2 1.000 0.555 0.555
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t2 ~
selfes_t1 1.000 0.982 0.982
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t1 ~~
d_slf_1 (fb_b) -0.077 0.034 -2.289 0.022 -0.251 -0.251
.selfes_par1_t1 ~~
.slf_1_2 (cv1b) 0.060 0.024 2.441 0.015 0.060 0.373
.selfes_par2_t1 ~~
.slf_2_2 (cv2b) 0.058 0.029 2.017 0.044 0.058 0.399
.selfes_par3_t1 ~~
.slf_3_2 (cv3b) 0.108 0.040 2.710 0.007 0.108 0.264
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.slfs_t2 0.000 0.000 0.000
d_slf_1 (d_n_) 0.166 0.040 4.158 0.000 0.398 0.398
slfs_t1 (wb__) 3.420 0.062 55.352 0.000 4.635 4.635
.slf_1_1 0.000 0.000 0.000
.slf_2_1 (m2b) -0.772 0.151 -5.096 0.000 -0.772 -0.827
.slf_3_1 (m3b) -0.966 0.187 -5.155 0.000 -0.966 -0.943
.slf_1_2 0.000 0.000 0.000
.slf_2_2 (m2b) -0.772 0.151 -5.096 0.000 -0.772 -0.815
.slf_3_2 (m3b) -0.966 0.187 -5.155 0.000 -0.966 -0.933
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.slfs_t2 0.000 0.000 0.000
d_slf_1 (d_v_) 0.174 0.036 4.817 0.000 1.000 1.000
slfs_t1 (wb__) 0.544 0.061 8.914 0.000 1.000 1.000
.slf_1_1 (rs1b) 0.160 0.024 6.757 0.000 0.160 0.227
.slf_2_1 (rs2b) 0.145 0.027 5.425 0.000 0.145 0.166
.slf_3_1 (rs3b) 0.410 0.040 10.133 0.000 0.410 0.391
.slf_1_2 (rs1b) 0.160 0.024 6.757 0.000 0.160 0.221
.slf_2_2 (rs2b) 0.145 0.027 5.425 0.000 0.145 0.161
.slf_3_2 (rs3b) 0.410 0.040 10.133 0.000 0.410 0.383Self-concept clarity: fitting multi-group models
# adapt latent change score model from above and add grouping factor in estimation (also add vectorized equality constraints to the model)
# configural invariance
mi_lcs_concept_group_config <- '
concept_t1 =~ 1*concept_par1_t1 + c("lamb2a", "lamb2b")*concept_par2_t1 + c("lamb3a", "lamb3b")*concept_par3_t1 # This specifies the measurement model for concept_t1
concept_t2 =~ 1*concept_par1_t2 + c("lamb2a", "lamb2b")*concept_par2_t2 + c("lamb3a", "lamb3b")*concept_par3_t2 # This specifies the measurement model for concept_t2 with the equality constrained factor loadings
concept_t2 ~ 1*concept_t1 # This parameter regresses concept_t2 perfectly on concept_t1
d_concept_1 =~ 1*concept_t2 # This defines the latent change score factor as measured perfectly by scores on concept_t2
concept_t2 ~ 0*1 # This line constrains the intercept of concept_t2 to 0
concept_t2 ~~ 0*concept_t2 # This fixes the variance of concept_t2 to 0
d_concept_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
concept_t1 ~ c("wb_int_a", "wb_int_b")*1 # This estimates the intercept of concept_t1
d_concept_1 ~~ c("d_var_a", "d_var_b")*d_concept_1 # This estimates the variance of the change scores
concept_t1 ~~ c("wb_var_a", "wb_var_b")*concept_t1 # This estimates the variance of the concept_t1
d_concept_1 ~~ c("fb_a", "fb_b")*concept_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
concept_par1_t1 ~~ c("cov1a", "cov1b")*concept_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
concept_par2_t1 ~~ c("cov2a", "cov2b")*concept_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
concept_par3_t1 ~~ c("cov3a", "cov3b")*concept_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
concept_par1_t1 ~~ c("res1a", "res1b")*concept_par1_t1 # This allows residual variance on indicator X1 at T1
concept_par2_t1 ~~ c("res2a", "res2b")*concept_par2_t1 # This allows residual variance on indicator X2 at T1
concept_par3_t1 ~~ c("res3a", "res3b")*concept_par3_t1 # This allows residual variance on indicator X3 at T1
concept_par1_t2 ~~ c("res1a", "res1b")*concept_par1_t2 # This allows residual variance on indicator X1 at T2
concept_par2_t2 ~~ c("res2a", "res2b")*concept_par2_t2 # This allows residual variance on indicator X2 at T2
concept_par3_t2 ~~ c("res3a", "res3b")*concept_par3_t2 # This allows residual variance on indicator X3 at T2
concept_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
concept_par2_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
concept_par3_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
concept_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
concept_par2_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
concept_par3_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
'
lcs_concept_group_12_config <- sem(mi_lcs_concept_group_config, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr',
fixed.x=FALSE, missing='fiml', group = "group")
# weak invariance
mi_lcs_concept_group_weak <- '
concept_t1 =~ 1*concept_par1_t1 + c("lamb2", "lamb2")*concept_par2_t1 + c("lamb3", "lamb3")*concept_par3_t1 # This specifies the measurement model for concept_t1
concept_t2 =~ 1*concept_par1_t2 + c("lamb2", "lamb2")*concept_par2_t2 + c("lamb3", "lamb3")*concept_par3_t2 # This specifies the measurement model for concept_t2 with the equality constrained factor loadings
concept_t2 ~ 1*concept_t1 # This parameter regresses concept_t2 perfectly on concept_t1
d_concept_1 =~ 1*concept_t2 # This defines the latent change score factor as measured perfectly by scores on concept_t2
concept_t2 ~ 0*1 # This line constrains the intercept of concept_t2 to 0
concept_t2 ~~ 0*concept_t2 # This fixes the variance of concept_t2 to 0
d_concept_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
concept_t1 ~ c("wb_int_a", "wb_int_b")*1 # This estimates the intercept of concept_t1
d_concept_1 ~~ c("d_var_a", "d_var_b")*d_concept_1 # This estimates the variance of the change scores
concept_t1 ~~ c("wb_var_a", "wb_var_b")*concept_t1 # This estimates the variance of the concept_t1
d_concept_1 ~~ c("fb_a", "fb_b")*concept_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
concept_par1_t1 ~~ c("cov1a", "cov1b")*concept_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
concept_par2_t1 ~~ c("cov2a", "cov2b")*concept_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
concept_par3_t1 ~~ c("cov3a", "cov3b")*concept_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
concept_par1_t1 ~~ c("res1a", "res1b")*concept_par1_t1 # This allows residual variance on indicator X1 at T1
concept_par2_t1 ~~ c("res2a", "res2b")*concept_par2_t1 # This allows residual variance on indicator X2 at T1
concept_par3_t1 ~~ c("res3a", "res3b")*concept_par3_t1 # This allows residual variance on indicator X3 at T1
concept_par1_t2 ~~ c("res1a", "res1b")*concept_par1_t2 # This allows residual variance on indicator X1 at T2
concept_par2_t2 ~~ c("res2a", "res2b")*concept_par2_t2 # This allows residual variance on indicator X2 at T2
concept_par3_t2 ~~ c("res3a", "res3b")*concept_par3_t2 # This allows residual variance on indicator X3 at T2
concept_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
concept_par2_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
concept_par3_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
concept_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
concept_par2_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
concept_par3_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
'
lcs_concept_group_12_weak <- sem(mi_lcs_concept_group_weak, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
# strong invariance
mi_lcs_concept_group_strong <- '
concept_t1 =~ 1*concept_par1_t1 + c("lamb2", "lamb2")*concept_par2_t1 + c("lamb3", "lamb3")*concept_par3_t1 # This specifies the measurement model for concept_t1
concept_t2 =~ 1*concept_par1_t2 + c("lamb2", "lamb2")*concept_par2_t2 + c("lamb3", "lamb3")*concept_par3_t2 # This specifies the measurement model for concept_t2 with the equality constrained factor loadings
concept_t2 ~ 1*concept_t1 # This parameter regresses concept_t2 perfectly on concept_t1
d_concept_1 =~ 1*concept_t2 # This defines the latent change score factor as measured perfectly by scores on concept_t2
concept_t2 ~ 0*1 # This line constrains the intercept of concept_t2 to 0
concept_t2 ~~ 0*concept_t2 # This fixes the variance of concept_t2 to 0
d_concept_1 ~ c("d_int", "d_int")*1 # This estimates the intercept of the change score
concept_t1 ~ c("wb_int", "wb_int")*1 # This estimates the intercept of concept_t1
d_concept_1 ~~ c("d_var_a", "d_var_b")*d_concept_1 # This estimates the variance of the change scores
concept_t1 ~~ c("wb_var_a", "wb_var_b")*concept_t1 # This estimates the variance of the concept_t1
d_concept_1 ~~ c("fb_a", "fb_b")*concept_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
concept_par1_t1 ~~ c("cov1a", "cov1b")*concept_par1_t2 # This allows residual covariance on indicator X1 across T1 and T2
concept_par2_t1 ~~ c("cov2a", "cov2b")*concept_par2_t2 # This allows residual covariance on indicator X2 across T1 and T2
concept_par3_t1 ~~ c("cov3a", "cov3b")*concept_par3_t2 # This allows residual covariance on indicator X3 across T1 and T2
concept_par1_t1 ~~ c("res1a", "res1b")*concept_par1_t1 # This allows residual variance on indicator X1 at T1
concept_par2_t1 ~~ c("res2a", "res2b")*concept_par2_t1 # This allows residual variance on indicator X2 at T1
concept_par3_t1 ~~ c("res3a", "res3b")*concept_par3_t1 # This allows residual variance on indicator X3 at T1
concept_par1_t2 ~~ c("res1a", "res1b")*concept_par1_t2 # This allows residual variance on indicator X1 at T2
concept_par2_t2 ~~ c("res2a", "res2b")*concept_par2_t2 # This allows residual variance on indicator X2 at T2
concept_par3_t2 ~~ c("res3a", "res3b")*concept_par3_t2 # This allows residual variance on indicator X3 at T2
concept_par1_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
concept_par2_t1 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T1
concept_par3_t1 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T1
concept_par1_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
concept_par2_t2 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T2
concept_par3_t2 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T2
'
lcs_concept_group_12_strong <- sem(mi_lcs_concept_group_strong, data=df_sbsa3_wide_wb %>% filter(group!="Group 3"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))Self-concept clarity: results
# model comparison tests for measurement invariance
lavTestLRT(lcs_concept_group_12_config, lcs_concept_group_12_weak, lcs_concept_group_12_strong)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
lcs_concept_group_12_config 24 3904.5 4020.6 30.375
lcs_concept_group_12_weak 26 3908.1 4016.4 37.911 7.7286 2
lcs_concept_group_12_strong 30 3905.3 3998.2 43.199 5.2123 4
Pr(>Chisq)
lcs_concept_group_12_config
lcs_concept_group_12_weak 0.02098 *
lcs_concept_group_12_strong 0.26620
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_concept_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_lcs_concept_group_12_weak <- broom::tidy(lcs_concept_group_12_weak, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_concept_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_lcs_concept_group_12_weak, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_concept_1 ~1 | d_int_a | 0.159 | 0.324 | 3.578 | 0.000 | |
| d_concept_1 ~1 | d_int_b | 0.080 | 0.181 | 1.903 | 0.057 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(lcs_concept_group_12_strong, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_concept_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_concept_1 ~1 | d_int | 0.117 | 0.237 | 3.826 | 0 |
| d_concept_1 ~1 | d_int | 0.117 | 0.265 | 3.826 | 0 |
# whole model (weak invariance)
summary(lcs_concept_group_12_weak, fit.measures=TRUE, standardized=TRUE, rsquare=F)lavaan 0.6.15 ended normally after 60 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 44
Number of equality constraints 16
Number of observations per group:
Group 1 175
Group 2 179
Number of missing patterns per group:
Group 1 3
Group 2 2
Model Test User Model:
Standard Scaled
Test Statistic 37.911 38.055
Degrees of freedom 26 26
P-value (Chi-square) 0.062 0.060
Scaling correction factor 0.996
Yuan-Bentler correction (Mplus variant)
Test statistic for each group:
Group 1 26.078 26.176
Group 2 11.834 11.878
Model Test Baseline Model:
Test statistic 1625.355 1381.998
Degrees of freedom 30 30
P-value 0.000 0.000
Scaling correction factor 1.176
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.993 0.991
Tucker-Lewis Index (TLI) 0.991 0.990
Robust Comparative Fit Index (CFI) 0.992
Robust Tucker-Lewis Index (TLI) 0.991
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -1926.027 -1926.027
Scaling correction factor 0.670
for the MLR correction
Loglikelihood unrestricted model (H1) -1907.072 -1907.072
Scaling correction factor 1.025
for the MLR correction
Akaike (AIC) 3908.054 3908.054
Bayesian (BIC) 4016.395 4016.395
Sample-size adjusted Bayesian (SABIC) 3927.567 3927.567
Root Mean Square Error of Approximation:
RMSEA 0.051 0.051
90 Percent confidence interval - lower 0.000 0.000
90 Percent confidence interval - upper 0.084 0.084
P-value H_0: RMSEA <= 0.050 0.451 0.445
P-value H_0: RMSEA >= 0.080 0.077 0.080
Robust RMSEA 0.054
90 Percent confidence interval - lower 0.000
90 Percent confidence interval - upper 0.089
P-value H_0: Robust RMSEA <= 0.050 0.400
P-value H_0: Robust RMSEA >= 0.080 0.118
Standardized Root Mean Square Residual:
SRMR 0.040 0.040
Parameter Estimates:
Standard errors Sandwich
Information bread Observed
Observed information based on Hessian
Group 1 [Group 1]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t1 =~
cnc_1_1 1.000 0.742 0.824
cnc_2_1 (lmb2) 1.215 0.049 24.807 0.000 0.902 0.880
cnc_3_1 (lmb3) 1.217 0.055 22.284 0.000 0.903 0.910
concept_t2 =~
cnc_1_2 1.000 0.764 0.832
cnc_2_2 (lmb2) 1.215 0.049 24.807 0.000 0.928 0.886
cnc_3_2 (lmb3) 1.217 0.055 22.284 0.000 0.930 0.914
d_concept_1 =~
cncpt_2 1.000 0.642 0.642
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t2 ~
concept_t1 1.000 0.971 0.971
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t1 ~~
d_cnc_1 (fb_a) -0.104 0.035 -2.975 0.003 -0.285 -0.285
.concept_par1_t1 ~~
.cnc_1_2 (cov1) 0.082 0.032 2.541 0.011 0.082 0.313
.concept_par2_t1 ~~
.cnc_2_2 (cov2) 0.090 0.031 2.886 0.004 0.090 0.379
.concept_par3_t1 ~~
.cnc_3_2 (cov3) 0.079 0.028 2.831 0.005 0.079 0.464
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.cncpt_2 0.000 0.000 0.000
d_cnc_1 (d_n_) 0.159 0.044 3.578 0.000 0.324 0.324
cncpt_1 (wb__) 3.043 0.067 45.621 0.000 4.100 4.100
.cnc_1_1 0.000 0.000 0.000
.cnc_2_1 (m2a) -0.590 0.164 -3.604 0.000 -0.590 -0.576
.cnc_3_1 (m3a) -0.528 0.181 -2.923 0.003 -0.528 -0.532
.cnc_1_2 0.000 0.000 0.000
.cnc_2_2 (m2a) -0.590 0.164 -3.604 0.000 -0.590 -0.563
.cnc_3_2 (m3a) -0.528 0.181 -2.923 0.003 -0.528 -0.519
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.cncpt_2 0.000 0.000 0.000
d_cnc_1 (d_v_) 0.240 0.056 4.310 0.000 1.000 1.000
cncpt_1 (wb__) 0.551 0.062 8.909 0.000 1.000 1.000
.cnc_1_1 (res1) 0.261 0.034 7.716 0.000 0.261 0.321
.cnc_2_1 (res2) 0.237 0.032 7.454 0.000 0.237 0.226
.cnc_3_1 (res3) 0.170 0.031 5.545 0.000 0.170 0.173
.cnc_1_2 (res1) 0.261 0.034 7.716 0.000 0.261 0.309
.cnc_2_2 (res2) 0.237 0.032 7.454 0.000 0.237 0.216
.cnc_3_2 (res3) 0.170 0.031 5.545 0.000 0.170 0.165
Group 2 [Group 2]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t1 =~
cnc_1_1 1.000 0.692 0.807
cnc_2_1 (lmb2) 1.215 0.049 24.807 0.000 0.840 0.854
cnc_3_1 (lmb3) 1.217 0.055 22.284 0.000 0.842 0.903
concept_t2 =~
cnc_1_2 1.000 0.703 0.812
cnc_2_2 (lmb2) 1.215 0.049 24.807 0.000 0.854 0.857
cnc_3_2 (lmb3) 1.217 0.055 22.284 0.000 0.856 0.906
d_concept_1 =~
cncpt_2 1.000 0.624 0.624
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t2 ~
concept_t1 1.000 0.984 0.984
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t1 ~~
d_cnc_1 (fb_b) -0.088 0.029 -3.026 0.002 -0.291 -0.291
.concept_par1_t1 ~~
.cnc_1_2 (cv1b) 0.075 0.027 2.760 0.006 0.075 0.293
.concept_par2_t1 ~~
.cnc_2_2 (cv2b) 0.080 0.031 2.618 0.009 0.080 0.304
.concept_par3_t1 ~~
.cnc_3_2 (cv3b) 0.057 0.030 1.905 0.057 0.057 0.352
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.cncpt_2 0.000 0.000 0.000
d_cnc_1 (d_n_) 0.080 0.042 1.903 0.057 0.181 0.181
cncpt_1 (wb__) 2.921 0.059 49.287 0.000 4.222 4.222
.cnc_1_1 0.000 0.000 0.000
.cnc_2_1 (m2b) -0.534 0.151 -3.527 0.000 -0.534 -0.543
.cnc_3_1 (m3b) -0.444 0.168 -2.642 0.008 -0.444 -0.476
.cnc_1_2 0.000 0.000 0.000
.cnc_2_2 (m2b) -0.534 0.151 -3.527 0.000 -0.534 -0.536
.cnc_3_2 (m3b) -0.444 0.168 -2.642 0.008 -0.444 -0.470
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.cncpt_2 0.000 0.000 0.000
d_cnc_1 (d_v_) 0.193 0.034 5.748 0.000 1.000 1.000
cncpt_1 (wb__) 0.478 0.056 8.534 0.000 1.000 1.000
.cnc_1_1 (rs1b) 0.256 0.025 10.385 0.000 0.256 0.349
.cnc_2_1 (rs2b) 0.263 0.035 7.426 0.000 0.263 0.271
.cnc_3_1 (rs3b) 0.161 0.031 5.257 0.000 0.161 0.185
.cnc_1_2 (rs1b) 0.256 0.025 10.385 0.000 0.256 0.341
.cnc_2_2 (rs2b) 0.263 0.035 7.426 0.000 0.263 0.265
.cnc_3_2 (rs3b) 0.161 0.031 5.257 0.000 0.161 0.180Adapt latent change score model from above (but without any moderation) and add grouping factor in estimation (adding vectorized equality constraints to the model step by step):
# create templates:
# configural invariance
trait_template_main_config <- '
trait_t1 =~ 1*ind01_t1 + c("lamb2a", "lamb2b")*ind02_t1 + c("lamb3a", "lamb3b")*ind03_t1 # This specifies the measurement model for trait_t1
trait_t2 =~ 1*ind01_t2 + c("lamb2a", "lamb2b")*ind02_t2 + c("lamb3a", "lamb3b")*ind03_t2 # This specifies the measurement model for trait_t2 with the equality constrained factor loadings
trait_t2 ~ 1*trait_t1 # This parameter regresses trait_t2 perfectly on trait_t1
d_trait_1 =~ 1*trait_t2 # This defines the latent change score factor as measured perfectly by scores on trait_t2
trait_t2 ~ 0*1 # This line constrains the intercept of trait_t2 to 0
trait_t2 ~~ 0*trait_t2 # This fixes the variance of trait_t2 to 0
d_trait_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
trait_t1 ~ c("b5_int_a", "b5_int_b")*1 # This estimates the intercept of trait_t1
d_trait_1 ~~ c("d_var_a", "d_var_b")*d_trait_1 # This estimates the variance of the change scores
trait_t1 ~~ c("b5_var_a", "b5_var_b")*trait_t1 # This estimates the variance of trait_t1
d_trait_1 ~~ c("fb_a", "fb_b")* trait_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
ind01_t1 ~~ c("cov1a", "cov1b")*ind01_t2 # This allows residual covariance on indicator X1 across T1 and T2
ind02_t1 ~~ c("cov2a", "cov2b")*ind02_t2 # This allows residual covariance on indicator X2 across T1 and T2
ind03_t1 ~~ c("cov3a", "cov3b")*ind03_t2 # This allows residual covariance on indicator X3 across T1 and T2
ind01_t1 ~~ c("res1a", "res1b")*ind01_t1 # This allows residual variance on indicator X1 at T1
ind02_t1 ~~ c("res2a", "res2b")*ind02_t1 # This allows residual variance on indicator X2 at T1
ind03_t1 ~~ c("res3a", "res3b")*ind03_t1 # This allows residual variance on indicator X3 at T1
ind01_t2 ~~ c("res1a", "res1b")*ind01_t2 # This allows residual variance on indicator X1 at T2
ind02_t2 ~~ c("res2a", "res2b")*ind02_t2 # This allows residual variance on indicator X2 at T2
ind03_t2 ~~ c("res3a", "res3b")*ind03_t2 # This allows residual variance on indicator X3 at T2
ind01_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
ind02_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
ind03_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
ind01_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
ind02_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
ind03_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
'
# weak invariance
trait_template_main_weak <- '
trait_t1 =~ 1*ind01_t1 + c("lamb2", "lamb2")*ind02_t1 + c("lamb3", "lamb3")*ind03_t1 # This specifies the measurement model for trait_t1
trait_t2 =~ 1*ind01_t2 + c("lamb2", "lamb2")*ind02_t2 + c("lamb3", "lamb3")*ind03_t2 # This specifies the measurement model for trait_t2 with the equality constrained factor loadings
trait_t2 ~ 1*trait_t1 # This parameter regresses trait_t2 perfectly on trait_t1
d_trait_1 =~ 1*trait_t2 # This defines the latent change score factor as measured perfectly by scores on trait_t2
trait_t2 ~ 0*1 # This line constrains the intercept of trait_t2 to 0
trait_t2 ~~ 0*trait_t2 # This fixes the variance of trait_t2 to 0
d_trait_1 ~ c("d_int_a", "d_int_b")*1 # This estimates the intercept of the change score
trait_t1 ~ c("b5_int_a", "b5_int_b")*1 # This estimates the intercept of trait_t1
d_trait_1 ~~ c("d_var_a", "d_var_b")*d_trait_1 # This estimates the variance of the change scores
trait_t1 ~~ c("b5_var_a", "b5_var_b")*trait_t1 # This estimates the variance of trait_t1
d_trait_1 ~~ c("fb_a", "fb_b")* trait_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
ind01_t1 ~~ c("cov1a", "cov1b")*ind01_t2 # This allows residual covariance on indicator X1 across T1 and T2
ind02_t1 ~~ c("cov2a", "cov2b")*ind02_t2 # This allows residual covariance on indicator X2 across T1 and T2
ind03_t1 ~~ c("cov3a", "cov3b")*ind03_t2 # This allows residual covariance on indicator X3 across T1 and T2
ind01_t1 ~~ c("res1a", "res1b")*ind01_t1 # This allows residual variance on indicator X1 at T1
ind02_t1 ~~ c("res2a", "res2b")*ind02_t1 # This allows residual variance on indicator X2 at T1
ind03_t1 ~~ c("res3a", "res3b")*ind03_t1 # This allows residual variance on indicator X3 at T1
ind01_t2 ~~ c("res1a", "res1b")*ind01_t2 # This allows residual variance on indicator X1 at T2
ind02_t2 ~~ c("res2a", "res2b")*ind02_t2 # This allows residual variance on indicator X2 at T2
ind03_t2 ~~ c("res3a", "res3b")*ind03_t2 # This allows residual variance on indicator X3 at T2
ind01_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
ind02_t1 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T1
ind03_t1 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T1
ind01_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
ind02_t2 ~ c("m2a", "m2b")*1 # This estimates the intercept of X2 at T2
ind03_t2 ~ c("m3a", "m3b")*1 # This estimates the intercept of X3 at T2
'
# strong invariance
trait_template_main_strong <- '
trait_t1 =~ 1*ind01_t1 + c("lamb2", "lamb2")*ind02_t1 + c("lamb3", "lamb3")*ind03_t1 # This specifies the measurement model for trait_t1
trait_t2 =~ 1*ind01_t2 + c("lamb2", "lamb2")*ind02_t2 + c("lamb3", "lamb3")*ind03_t2 # This specifies the measurement model for trait_t2 with the equality constrained factor loadings
trait_t2 ~ 1*trait_t1 # This parameter regresses trait_t2 perfectly on trait_t1
d_trait_1 =~ 1*trait_t2 # This defines the latent change score factor as measured perfectly by scores on trait_t2
trait_t2 ~ 0*1 # This line constrains the intercept of trait_t2 to 0
trait_t2 ~~ 0*trait_t2 # This fixes the variance of trait_t2 to 0
d_trait_1 ~ c("d_int", "d_int")*1 # This estimates the intercept of the change score
trait_t1 ~ c("b5_int", "b5_int")*1 # This estimates the intercept of trait_t1
d_trait_1 ~~ c("d_var_a", "d_var_b")*d_trait_1 # This estimates the variance of the change scores
trait_t1 ~~ c("b5_var_a", "b5_var_b")*trait_t1 # This estimates the variance of trait_t1
d_trait_1 ~~ c("fb_a", "fb_b")* trait_t1 # This estimates the self-feedback parameter, as a covariance! -> therefore, the interpretation of the change score
ind01_t1 ~~ c("cov1a", "cov1b")*ind01_t2 # This allows residual covariance on indicator X1 across T1 and T2
ind02_t1 ~~ c("cov2a", "cov2b")*ind02_t2 # This allows residual covariance on indicator X2 across T1 and T2
ind03_t1 ~~ c("cov3a", "cov3b")*ind03_t2 # This allows residual covariance on indicator X3 across T1 and T2
ind01_t1 ~~ c("res1a", "res1b")*ind01_t1 # This allows residual variance on indicator X1 at T1
ind02_t1 ~~ c("res2a", "res2b")*ind02_t1 # This allows residual variance on indicator X2 at T1
ind03_t1 ~~ c("res3a", "res3b")*ind03_t1 # This allows residual variance on indicator X3 at T1
ind01_t2 ~~ c("res1a", "res1b")*ind01_t2 # This allows residual variance on indicator X1 at T2
ind02_t2 ~~ c("res2a", "res2b")*ind02_t2 # This allows residual variance on indicator X2 at T2
ind03_t2 ~~ c("res3a", "res3b")*ind03_t2 # This allows residual variance on indicator X3 at T2
ind01_t1 ~ 0*1 # This constrains the intercept of X1 to 0 at T1
ind02_t1 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T1
ind03_t1 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T1
ind01_t2 ~ 0*1 # This constrains the intercept of X1 to 0 at T2
ind02_t2 ~ c("m2", "m2")*1 # This estimates the intercept of X2 at T2
ind03_t2 ~ c("m3", "m3")*1 # This estimates the intercept of X3 at T2
'
# strict invariance -> not really needed for this analysis
# loop across 5 traits
for (i in 1:5) {
item_nrs = b5_vars[[i]][[1]]
short_name = str_trunc(names(b5_vars)[i], 5, ellipsis = "")
# configural - current
template_filled_config_current <- str_replace_all(trait_template_main_config,
c("trait" = short_name,
"ind01" = paste0(short_name, "_curr_par1"),
"ind02" = paste0(short_name, "_curr_par2"),
"ind03" = paste0(short_name, "_curr_par3")))
trait_main_fit_config_current <- lavaan(template_filled_config_current,
data = df_sbsa3_wide_pers %>% filter(group!="Group 3"),
estimator='mlr', fixed.x=FALSE, missing='fiml', group = "group")
eval(call("<-", as.name(paste0("lcs_", short_name, "_main_config_group_12")), template_filled_config_current))
eval(call("<-", as.name(paste0("fit_lcs_", short_name, "_main_config_group_12")), trait_main_fit_config_current))
# weak - current
template_filled_weak_current <- str_replace_all(trait_template_main_weak,
c("trait" = short_name,
"ind01" = paste0(short_name, "_curr_par1"),
"ind02" = paste0(short_name, "_curr_par2"),
"ind03" = paste0(short_name, "_curr_par3")))
trait_main_fit_weak_current <- lavaan(template_filled_weak_current,
data = df_sbsa3_wide_pers %>% filter(group!="Group 3"),
estimator='mlr', fixed.x=FALSE, missing='fiml', group = "group", group.equal = "loadings")
eval(call("<-", as.name(paste0("lcs_", short_name, "_main_weak_group_12")), template_filled_weak_current))
eval(call("<-", as.name(paste0("fit_lcs_", short_name, "_main_weak_group_12")), trait_main_fit_weak_current))
# strong - current
template_filled_strong_current <- str_replace_all(trait_template_main_strong,
c("trait" = short_name,
"ind01" = paste0(short_name, "_curr_par1"),
"ind02" = paste0(short_name, "_curr_par2"),
"ind03" = paste0(short_name, "_curr_par3")))
trait_main_fit_strong_current <- lavaan(template_filled_strong_current,
data = df_sbsa3_wide_pers %>% filter(group!="Group 3"),
estimator='mlr', fixed.x=FALSE, missing='fiml', group = "group", group.equal = c("intercepts", "loadings"))
eval(call("<-", as.name(paste0("lcs_", short_name, "_main_strong_group_12")), template_filled_strong_current))
eval(call("<-", as.name(paste0("fit_lcs_", short_name, "_main_strong_group_12")), trait_main_fit_strong_current))
}Extraversion (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_extra_main_config_group_12, fit_lcs_extra_main_weak_group_12, fit_lcs_extra_main_strong_group_12)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_extra_main_config_group_12 24 3307.9 3423.9 24.541
fit_lcs_extra_main_weak_group_12 26 3307.4 3415.7 28.071 3.7366 2
fit_lcs_extra_main_strong_group_12 30 3302.9 3395.7 31.554 3.4568 4
Pr(>Chisq)
fit_lcs_extra_main_config_group_12
fit_lcs_extra_main_weak_group_12 0.1544
fit_lcs_extra_main_strong_group_12 0.4845# show model with varying latent change parameters
# -> key parameter is "d_extra_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_main_lcs_extra_group_12 <- broom::tidy(fit_lcs_extra_main_weak_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_extra_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_main_lcs_extra_group_12, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_extra_1 ~1 | d_int_a | 0.074 | 0.37 | 2.775 | 0.006 | |
| d_extra_1 ~1 | d_int_b | 0.088 | 0.29 | 2.739 | 0.006 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_extra_main_strong_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_extra_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_extra_1 ~1 | d_int | 0.08 | 0.398 | 3.831 | 0 |
| d_extra_1 ~1 | d_int | 0.08 | 0.262 | 3.831 | 0 |
# model fit
kable(broom::glance(fit_lcs_extra_main_weak_group_12) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 354 | 28 | 3307.394 | 3415.734 | 0.999 | 0.021 | 0.047 |
Agreeableness (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_agree_main_config_group_12, fit_lcs_agree_main_weak_group_12, fit_lcs_agree_main_strong_group_12)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_agree_main_config_group_12 24 3161.8 3277.9 39.952
fit_lcs_agree_main_weak_group_12 26 3158.5 3266.8 40.567 0.5738 2
fit_lcs_agree_main_strong_group_12 30 3156.4 3249.2 46.473 6.0557 4
Pr(>Chisq)
fit_lcs_agree_main_config_group_12
fit_lcs_agree_main_weak_group_12 0.7506
fit_lcs_agree_main_strong_group_12 0.1950# show model with varying latent change parameters
# -> key parameter is "d_agree_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_main_lcs_agree_group_12 <- broom::tidy(fit_lcs_agree_main_weak_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_agree_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_main_lcs_agree_group_12, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_agree_1 ~1 | d_int_a | 0.051 | 0.176 | 1.727 | 0.084 | |
| d_agree_1 ~1 | d_int_b | 0.047 | 0.168 | 1.623 | 0.105 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_agree_main_strong_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_agree_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_agree_1 ~1 | d_int | 0.051 | 0.174 | 2.427 | 0.015 |
| d_agree_1 ~1 | d_int | 0.051 | 0.181 | 2.427 | 0.015 |
# model fit
kable(broom::glance(fit_lcs_agree_main_weak_group_12) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 354 | 28 | 3158.464 | 3266.804 | 0.986 | 0.056 | 0.052 |
Conscientiousness (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_consc_main_config_group_12, fit_lcs_consc_main_weak_group_12, fit_lcs_consc_main_strong_group_12)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_consc_main_config_group_12 24 3310.7 3426.8 27.774
fit_lcs_consc_main_weak_group_12 26 3307.5 3415.8 28.559 0.9105 2
fit_lcs_consc_main_strong_group_12 30 3302.7 3395.6 31.826 3.2967 4
Pr(>Chisq)
fit_lcs_consc_main_config_group_12
fit_lcs_consc_main_weak_group_12 0.6343
fit_lcs_consc_main_strong_group_12 0.5094# show model with varying latent change parameters
# -> key parameter is "d_consc_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_main_lcs_consc_group_12 <- broom::tidy(fit_lcs_consc_main_weak_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_consc_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_main_lcs_consc_group_12, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_consc_1 ~1 | d_int_a | 0.099 | 0.326 | 2.925 | 0.003 | |
| d_consc_1 ~1 | d_int_b | 0.068 | 0.276 | 2.365 | 0.018 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_consc_main_strong_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_consc_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_consc_1 ~1 | d_int | 0.082 | 0.269 | 3.719 | 0 |
| d_consc_1 ~1 | d_int | 0.082 | 0.329 | 3.719 | 0 |
# model fit
kable(broom::glance(fit_lcs_consc_main_weak_group_12) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 354 | 28 | 3307.458 | 3415.798 | 0.998 | 0.024 | 0.03 |
Neuroticism (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_neuro_main_config_group_12, fit_lcs_neuro_main_weak_group_12, fit_lcs_neuro_main_strong_group_12)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_neuro_main_config_group_12 24 3307.8 3423.9 33.029
fit_lcs_neuro_main_weak_group_12 26 3303.9 3412.2 33.086 0.0598 2
fit_lcs_neuro_main_strong_group_12 30 3299.2 3392.1 36.416 3.3577 4
Pr(>Chisq)
fit_lcs_neuro_main_config_group_12
fit_lcs_neuro_main_weak_group_12 0.9705
fit_lcs_neuro_main_strong_group_12 0.4998# show model with varying latent change parameters
# -> key parameter is "d_neuro_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_main_lcs_neuro_group_12 <- broom::tidy(fit_lcs_neuro_main_weak_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_neuro_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_main_lcs_neuro_group_12, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_neuro_1 ~1 | d_int_a | -0.052 | -0.145 | -1.436 | 0.151 | |
| d_neuro_1 ~1 | d_int_b | -0.128 | -0.462 | -4.271 | 0.000 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_neuro_main_strong_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_neuro_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_neuro_1 ~1 | d_int | -0.098 | -0.270 | -4.066 | 0 |
| d_neuro_1 ~1 | d_int | -0.098 | -0.349 | -4.066 | 0 |
# model fit
kable(broom::glance(fit_lcs_neuro_main_weak_group_12) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 354 | 28 | 3303.86 | 3412.2 | 0.996 | 0.039 | 0.023 |
Openness (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_openn_main_config_group_12, fit_lcs_openn_main_weak_group_12, fit_lcs_openn_main_strong_group_12)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_openn_main_config_group_12 24 3369.4 3485.5 30.630
fit_lcs_openn_main_weak_group_12 26 3365.5 3473.9 30.784 0.14398 2
fit_lcs_openn_main_strong_group_12 30 3359.3 3452.1 32.526 1.71588 4
Pr(>Chisq)
fit_lcs_openn_main_config_group_12
fit_lcs_openn_main_weak_group_12 0.9305
fit_lcs_openn_main_strong_group_12 0.7878# show model with varying latent change parameters
# -> key parameter is "d_openn_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 1, b = Group 2)
params_main_lcs_openn_group_12 <- broom::tidy(fit_lcs_openn_main_weak_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_openn_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_main_lcs_openn_group_12, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_openn_1 ~1 | d_int_a | 0.022 | 0.110 | 0.821 | 0.412 | |
| d_openn_1 ~1 | d_int_b | 0.044 | 0.221 | 1.798 | 0.072 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_openn_main_strong_group_12, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_openn_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_openn_1 ~1 | d_int | 0.034 | 0.167 | 1.797 | 0.072 |
| d_openn_1 ~1 | d_int | 0.034 | 0.171 | 1.797 | 0.072 |
# model fit
kable(broom::glance(fit_lcs_openn_main_weak_group_12) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 354 | 28 | 3365.527 | 3473.867 | 0.996 | 0.032 | 0.036 |
In Study 3, we can only examine this in Group 1, so no multi-group comparison of ideal personality possible.
Life satisfaction: fitting multi-group models
# adapt latent change score model from above and add grouping factor in estimation (also add vectorized equality constraints to the model)
# configural invariance -> model template created above
lcs_swls_group_23_config <- sem(mi_lcs_swls_group_config, data=df_sbsa3_wide_wb %>% filter(group!="Group 1"), estimator='mlr',
fixed.x=FALSE, missing='fiml', group = "group")
# weak invariance -> model template created above
lcs_swls_group_23_weak <- sem(mi_lcs_swls_group_weak, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
# strong invariance -> model template created above
lcs_swls_group_23_strong <- sem(mi_lcs_swls_group_strong, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))Life satisfaction: results
# model comparison tests for measurement invariance
lavTestLRT(lcs_swls_group_23_config, lcs_swls_group_23_weak, lcs_swls_group_23_strong)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff Pr(>Chisq)
lcs_swls_group_23_config 50 6785.3 6932.4 85.341
lcs_swls_group_23_weak 53 6779.9 6915.4 85.974 0.6112 3 0.89386
lcs_swls_group_23_strong 58 6785.8 6902.0 101.870 15.7264 5 0.00767
lcs_swls_group_23_config
lcs_swls_group_23_weak
lcs_swls_group_23_strong **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_swls_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_lcs_swls_group_23_weak <- broom::tidy(lcs_swls_group_23_weak, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_swls_1 ~1 ")) %>% mutate(sig_diff = "**")
kable(params_lcs_swls_group_23_weak, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_swls_1 ~1 | d_int_a | 0.255 | 0.511 | 4.850 | 0.00 | ** |
| d_swls_1 ~1 | d_int_b | 0.019 | 0.034 | 0.358 | 0.72 | ** |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(lcs_swls_group_23_strong, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_swls_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_swls_1 ~1 | d_int | 0.14 | 0.275 | 3.377 | 0.001 |
| d_swls_1 ~1 | d_int | 0.14 | 0.241 | 3.377 | 0.001 |
# whole model (weak invariance)
summary(lcs_swls_group_23_weak, fit.measures=TRUE, standardized=TRUE, rsquare=F)lavaan 0.6.15 ended normally after 53 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 58
Number of equality constraints 23
Number of observations per group:
Group 2 179
Group 3 176
Number of missing patterns per group:
Group 2 2
Group 3 2
Model Test User Model:
Standard Scaled
Test Statistic 85.974 75.731
Degrees of freedom 53 53
P-value (Chi-square) 0.003 0.022
Scaling correction factor 1.135
Yuan-Bentler correction (Mplus variant)
Test statistic for each group:
Group 2 53.793 47.384
Group 3 32.181 28.347
Model Test Baseline Model:
Test statistic 1932.415 1344.826
Degrees of freedom 56 56
P-value 0.000 0.000
Scaling correction factor 1.437
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.982 0.982
Tucker-Lewis Index (TLI) 0.981 0.981
Robust Comparative Fit Index (CFI) 0.986
Robust Tucker-Lewis Index (TLI) 0.985
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -3354.947 -3354.947
Scaling correction factor 0.757
for the MLR correction
Loglikelihood unrestricted model (H1) -3311.960 -3311.960
Scaling correction factor 1.183
for the MLR correction
Akaike (AIC) 6779.893 6779.893
Bayesian (BIC) 6915.417 6915.417
Sample-size adjusted Bayesian (SABIC) 6804.382 6804.382
Root Mean Square Error of Approximation:
RMSEA 0.059 0.049
90 Percent confidence interval - lower 0.035 0.022
90 Percent confidence interval - upper 0.081 0.071
P-value H_0: RMSEA <= 0.050 0.242 0.502
P-value H_0: RMSEA >= 0.080 0.063 0.009
Robust RMSEA 0.055
90 Percent confidence interval - lower 0.021
90 Percent confidence interval - upper 0.081
P-value H_0: Robust RMSEA <= 0.050 0.371
P-value H_0: Robust RMSEA >= 0.080 0.061
Standardized Root Mean Square Residual:
SRMR 0.040 0.040
Parameter Estimates:
Standard errors Sandwich
Information bread Observed
Observed information based on Hessian
Group 1 [Group 2]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t1 =~
s06_01_ 1.000 0.968 0.815
s06_02_ (lmb2) 0.956 0.040 23.954 0.000 0.925 0.788
s06_03_ (lmb3) 1.084 0.050 21.826 0.000 1.050 0.868
s06_04_ (lmb4) 0.906 0.057 15.926 0.000 0.877 0.718
swls_t2 =~
s06_01_ 1.000 0.975 0.817
s06_02_ (lmb2) 0.956 0.040 23.954 0.000 0.932 0.790
s06_03_ (lmb3) 1.084 0.050 21.826 0.000 1.058 0.870
s06_04_ (lmb4) 0.906 0.057 15.926 0.000 0.884 0.721
d_swls_1 =~
swls_t2 1.000 0.511 0.511
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t2 ~
swls_t1 1.000 0.993 0.993
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t1 ~~
d_swl_1 (fb_a) -0.117 0.050 -2.334 0.020 -0.243 -0.243
.sw06_01_t1 ~~
.s06_01_ (cov1) 0.190 0.058 3.268 0.001 0.190 0.400
.sw06_02_t1 ~~
.s06_02_ (cov2) 0.206 0.061 3.395 0.001 0.206 0.395
.sw06_03_t1 ~~
.s06_03_ (cov3) 0.016 0.054 0.292 0.770 0.016 0.044
.sw06_04_t1 ~~
.s06_04_ (cov4) 0.199 0.074 2.674 0.007 0.199 0.275
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.swls_t2 0.000 0.000 0.000
d_swl_1 (d_n_) 0.255 0.052 4.850 0.000 0.511 0.511
swls_t1 (wb__) 2.546 0.087 29.366 0.000 2.630 2.630
.s06_01_ 0.000 0.000 0.000
.s06_02_ (m2a) 0.278 0.123 2.266 0.023 0.278 0.237
.s06_03_ (m3a) 0.062 0.144 0.431 0.666 0.062 0.051
.s06_04_ (m4a) 0.437 0.165 2.650 0.008 0.437 0.358
.s06_01_ 0.000 0.000 0.000
.s06_02_ (m2a) 0.278 0.123 2.266 0.023 0.278 0.236
.s06_03_ (m3a) 0.062 0.144 0.431 0.666 0.062 0.051
.s06_04_ (m4a) 0.437 0.165 2.650 0.008 0.437 0.356
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.swls_t2 0.000 0.000 0.000
d_swl_1 (d_v_) 0.248 0.055 4.490 0.000 1.000 1.000
swls_t1 (wb__) 0.937 0.104 9.029 0.000 1.000 1.000
.s06_01_ (res1) 0.474 0.063 7.529 0.000 0.474 0.336
.s06_02_ (res2) 0.523 0.061 8.561 0.000 0.523 0.379
.s06_03_ (res3) 0.361 0.055 6.528 0.000 0.361 0.247
.s06_04_ (res4) 0.722 0.078 9.252 0.000 0.722 0.484
.s06_01_ (res1) 0.474 0.063 7.529 0.000 0.474 0.333
.s06_02_ (res2) 0.523 0.061 8.561 0.000 0.523 0.376
.s06_03_ (res3) 0.361 0.055 6.528 0.000 0.361 0.244
.s06_04_ (res4) 0.722 0.078 9.252 0.000 0.722 0.480
Group 2 [Group 3]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t1 =~
s06_01_ 1.000 0.983 0.812
s06_02_ (lmb2) 0.956 0.040 23.954 0.000 0.940 0.802
s06_03_ (lmb3) 1.084 0.050 21.826 0.000 1.066 0.887
s06_04_ (lmb4) 0.906 0.057 15.926 0.000 0.891 0.718
swls_t2 =~
s06_01_ 1.000 1.020 0.822
s06_02_ (lmb2) 0.956 0.040 23.954 0.000 0.975 0.812
s06_03_ (lmb3) 1.084 0.050 21.826 0.000 1.106 0.894
s06_04_ (lmb4) 0.906 0.057 15.926 0.000 0.925 0.731
d_swls_1 =~
swls_t2 1.000 0.556 0.556
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t2 ~
swls_t1 1.000 0.964 0.964
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
swls_t1 ~~
d_swl_1 (fb_b) -0.124 0.069 -1.804 0.071 -0.222 -0.222
.sw06_01_t1 ~~
.s06_01_ (cv1b) 0.163 0.062 2.615 0.009 0.163 0.327
.sw06_02_t1 ~~
.s06_02_ (cv2b) 0.181 0.065 2.780 0.005 0.181 0.369
.sw06_03_t1 ~~
.s06_03_ (cv3b) 0.090 0.052 1.728 0.084 0.090 0.291
.sw06_04_t1 ~~
.s06_04_ (cv4b) 0.404 0.091 4.429 0.000 0.404 0.541
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.swls_t2 0.000 0.000 0.000
d_swl_1 (d_n_) 0.019 0.054 0.358 0.720 0.034 0.034
swls_t1 (wb__) 2.914 0.089 32.652 0.000 2.963 2.963
.s06_01_ 0.000 0.000 0.000
.s06_02_ (m2b) 0.257 0.141 1.831 0.067 0.257 0.220
.s06_03_ (m3b) -0.048 0.163 -0.291 0.771 -0.048 -0.040
.s06_04_ (m4b) 0.406 0.186 2.181 0.029 0.406 0.327
.s06_01_ 0.000 0.000 0.000
.s06_02_ (m2b) 0.257 0.141 1.831 0.067 0.257 0.214
.s06_03_ (m3b) -0.048 0.163 -0.291 0.771 -0.048 -0.038
.s06_04_ (m4b) 0.406 0.186 2.181 0.029 0.406 0.321
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.swls_t2 0.000 0.000 0.000
d_swl_1 (d_v_) 0.322 0.105 3.053 0.002 1.000 1.000
swls_t1 (wb__) 0.967 0.106 9.080 0.000 1.000 1.000
.s06_01_ (rs1b) 0.498 0.071 7.022 0.000 0.498 0.340
.s06_02_ (rs2b) 0.491 0.061 8.053 0.000 0.491 0.357
.s06_03_ (rs3b) 0.309 0.063 4.919 0.000 0.309 0.214
.s06_04_ (rs4b) 0.747 0.096 7.800 0.000 0.747 0.485
.s06_01_ (rs1b) 0.498 0.071 7.022 0.000 0.498 0.324
.s06_02_ (rs2b) 0.491 0.061 8.053 0.000 0.491 0.341
.s06_03_ (rs3b) 0.309 0.063 4.919 0.000 0.309 0.201
.s06_04_ (rs4b) 0.747 0.096 7.800 0.000 0.747 0.466Meaning in life: fitting multi-group models
# adapt latent change score model from above and add grouping factor in estimation (also add vectorized equality constraints to the model)
# configural invariance -> model template created above
lcs_meaning_group_23_config <- sem(mi_lcs_meaning_group_config, data=df_sbsa3_wide_wb %>% filter(group!="Group 1"), estimator='mlr',
fixed.x=FALSE, missing='fiml', group = "group")
# weak invariance -> model template created above
lcs_meaning_group_23_weak <- sem(mi_lcs_meaning_group_weak, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
# strong invariance -> model template created above
lcs_meaning_group_23_strong <- sem(mi_lcs_meaning_group_strong, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))Meaning in life: results
# model comparison tests for measurement invariance
lavTestLRT(lcs_meaning_group_23_config, lcs_meaning_group_23_weak, lcs_meaning_group_23_strong)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
lcs_meaning_group_23_config 24 5066.7 5182.9 34.651
lcs_meaning_group_23_weak 26 5063.3 5171.7 35.210 0.6420 2
lcs_meaning_group_23_strong 30 5059.6 5152.6 39.545 4.2488 4
Pr(>Chisq)
lcs_meaning_group_23_config
lcs_meaning_group_23_weak 0.7254
lcs_meaning_group_23_strong 0.3734# show model with varying latent change parameters
# -> key parameter is "d_meaning_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_lcs_meaning_group_23_weak <- broom::tidy(lcs_meaning_group_23_weak, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_meaning_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_lcs_meaning_group_23_weak, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_meaning_1 ~1 | d_int_a | 0.114 | 0.196 | 1.994 | 0.046 | |
| d_meaning_1 ~1 | d_int_b | 0.084 | 0.146 | 1.578 | 0.115 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(lcs_meaning_group_23_strong, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_meaning_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_meaning_1 ~1 | d_int | 0.097 | 0.165 | 2.449 | 0.014 |
| d_meaning_1 ~1 | d_int | 0.097 | 0.168 | 2.449 | 0.014 |
# whole model (weak invariance)
summary(lcs_meaning_group_23_weak, fit.measures=TRUE, standardized=TRUE, rsquare=F)lavaan 0.6.15 ended normally after 57 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 44
Number of equality constraints 16
Number of observations per group:
Group 2 179
Group 3 176
Number of missing patterns per group:
Group 2 2
Group 3 2
Model Test User Model:
Standard Scaled
Test Statistic 35.210 34.558
Degrees of freedom 26 26
P-value (Chi-square) 0.107 0.122
Scaling correction factor 1.019
Yuan-Bentler correction (Mplus variant)
Test statistic for each group:
Group 2 14.514 14.245
Group 3 20.697 20.313
Model Test Baseline Model:
Test statistic 1623.316 1398.857
Degrees of freedom 30 30
P-value 0.000 0.000
Scaling correction factor 1.160
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.994 0.994
Tucker-Lewis Index (TLI) 0.993 0.993
Robust Comparative Fit Index (CFI) 0.995
Robust Tucker-Lewis Index (TLI) 0.994
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -2503.651 -2503.651
Scaling correction factor 0.701
for the MLR correction
Loglikelihood unrestricted model (H1) -2486.045 -2486.045
Scaling correction factor 1.062
for the MLR correction
Akaike (AIC) 5063.301 5063.301
Bayesian (BIC) 5171.720 5171.720
Sample-size adjusted Bayesian (SABIC) 5082.892 5082.892
Root Mean Square Error of Approximation:
RMSEA 0.045 0.043
90 Percent confidence interval - lower 0.000 0.000
90 Percent confidence interval - upper 0.079 0.078
P-value H_0: RMSEA <= 0.050 0.562 0.590
P-value H_0: RMSEA >= 0.080 0.045 0.038
Robust RMSEA 0.043
90 Percent confidence interval - lower 0.000
90 Percent confidence interval - upper 0.082
P-value H_0: Robust RMSEA <= 0.050 0.570
P-value H_0: Robust RMSEA >= 0.080 0.060
Standardized Root Mean Square Residual:
SRMR 0.037 0.037
Parameter Estimates:
Standard errors Sandwich
Information bread Observed
Observed information based on Hessian
Group 1 [Group 2]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t1 =~
mnn_1_1 1.000 0.963 0.750
mnn_2_1 (lmb2) 1.018 0.042 24.165 0.000 0.981 0.842
mnn_3_1 (lmb3) 1.206 0.053 22.735 0.000 1.161 0.900
meaning_t2 =~
mnn_1_2 1.000 0.942 0.743
mnn_2_2 (lmb2) 1.018 0.042 24.165 0.000 0.960 0.836
mnn_3_2 (lmb3) 1.206 0.053 22.735 0.000 1.136 0.896
d_meaning_1 =~
mnng_t2 1.000 0.621 0.621
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t2 ~
meaning_t1 1.000 1.022 1.022
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t1 ~~
d_mnn_1 (fb_a) -0.191 0.081 -2.364 0.018 -0.339 -0.339
.meaning_par1_t1 ~~
.mnn_1_2 (cov1) 0.420 0.090 4.678 0.000 0.420 0.581
.meaning_par2_t1 ~~
.mnn_2_2 (cov2) 0.120 0.048 2.510 0.012 0.120 0.304
.meaning_par3_t1 ~~
.mnn_3_2 (cov3) 0.048 0.054 0.889 0.374 0.048 0.153
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.mnng_t2 0.000 0.000 0.000
d_mnn_1 (d_n_) 0.114 0.057 1.994 0.046 0.196 0.196
mnng_t1 (wb__) 3.352 0.091 36.946 0.000 3.481 3.481
.mnn_1_1 0.000 0.000 0.000
.mnn_2_1 (m2a) 0.304 0.158 1.916 0.055 0.304 0.260
.mnn_3_1 (m3a) -0.245 0.195 -1.256 0.209 -0.245 -0.190
.mnn_1_2 0.000 0.000 0.000
.mnn_2_2 (m2a) 0.304 0.158 1.916 0.055 0.304 0.265
.mnn_3_2 (m3a) -0.245 0.195 -1.256 0.209 -0.245 -0.193
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.mnng_t2 0.000 0.000 0.000
d_mnn_1 (d_v_) 0.342 0.088 3.882 0.000 1.000 1.000
mnng_t1 (wb__) 0.928 0.133 6.972 0.000 1.000 1.000
.mnn_1_1 (res1) 0.722 0.087 8.292 0.000 0.722 0.438
.mnn_2_1 (res2) 0.396 0.053 7.474 0.000 0.396 0.292
.mnn_3_1 (res3) 0.316 0.058 5.469 0.000 0.316 0.190
.mnn_1_2 (res1) 0.722 0.087 8.292 0.000 0.722 0.449
.mnn_2_2 (res2) 0.396 0.053 7.474 0.000 0.396 0.301
.mnn_3_2 (res3) 0.316 0.058 5.469 0.000 0.316 0.197
Group 2 [Group 3]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t1 =~
mnn_1_1 1.000 1.040 0.790
mnn_2_1 (lmb2) 1.018 0.042 24.165 0.000 1.059 0.876
mnn_3_1 (lmb3) 1.206 0.053 22.735 0.000 1.253 0.946
meaning_t2 =~
mnn_1_2 1.000 1.073 0.799
mnn_2_2 (lmb2) 1.018 0.042 24.165 0.000 1.092 0.882
mnn_3_2 (lmb3) 1.206 0.053 22.735 0.000 1.293 0.949
d_meaning_1 =~
mnng_t2 1.000 0.536 0.536
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t2 ~
meaning_t1 1.000 0.969 0.969
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
meaning_t1 ~~
d_mnn_1 (fb_b) -0.130 0.063 -2.083 0.037 -0.218 -0.218
.meaning_par1_t1 ~~
.mnn_1_2 (cv1b) 0.358 0.087 4.115 0.000 0.358 0.551
.meaning_par2_t1 ~~
.mnn_2_2 (cv2b) 0.104 0.045 2.311 0.021 0.104 0.304
.meaning_par3_t1 ~~
.mnn_3_2 (cv3b) -0.006 0.046 -0.133 0.894 -0.006 -0.033
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.mnng_t2 0.000 0.000 0.000
d_mnn_1 (d_n_) 0.084 0.053 1.578 0.115 0.146 0.146
mnng_t1 (wb__) 3.485 0.095 36.768 0.000 3.352 3.352
.mnn_1_1 0.000 0.000 0.000
.mnn_2_1 (m2b) 0.405 0.163 2.485 0.013 0.405 0.335
.mnn_3_1 (m3b) -0.226 0.196 -1.154 0.249 -0.226 -0.170
.mnn_1_2 0.000 0.000 0.000
.mnn_2_2 (m2b) 0.405 0.163 2.485 0.013 0.405 0.327
.mnn_3_2 (m3b) -0.226 0.196 -1.154 0.249 -0.226 -0.166
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.mnng_t2 0.000 0.000 0.000
d_mnn_1 (d_v_) 0.330 0.060 5.487 0.000 1.000 1.000
mnng_t1 (wb__) 1.081 0.147 7.330 0.000 1.000 1.000
.mnn_1_1 (rs1b) 0.650 0.088 7.378 0.000 0.650 0.376
.mnn_2_1 (rs2b) 0.341 0.045 7.599 0.000 0.341 0.233
.mnn_3_1 (rs3b) 0.185 0.049 3.779 0.000 0.185 0.106
.mnn_1_2 (rs1b) 0.650 0.088 7.378 0.000 0.650 0.361
.mnn_2_2 (rs2b) 0.341 0.045 7.599 0.000 0.341 0.222
.mnn_3_2 (rs3b) 0.185 0.049 3.779 0.000 0.185 0.100Self-esteem: fitting multi-group models
# adapt latent change score model from above and add grouping factor in estimation (also add vectorized equality constraints to the model)
# configural invariance -> model template created above
lcs_selfes_group_23_config <- sem(mi_lcs_selfes_group_config, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr',
fixed.x=FALSE, missing='fiml', group = "group")
# weak invariance -> model template created above
lcs_selfes_group_23_weak <- sem(mi_lcs_selfes_group_weak, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
# strong invariance -> model template created above
lcs_selfes_group_23_strong <- sem(mi_lcs_selfes_group_strong, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))Self-esteem: results
# model comparison tests for measurement invariance
lavTestLRT(lcs_selfes_group_23_config, lcs_selfes_group_23_weak, lcs_selfes_group_23_strong)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
lcs_selfes_group_23_config 24 3712.6 3828.7 23.270
lcs_selfes_group_23_weak 26 3711.2 3819.6 25.907 2.5162 2
lcs_selfes_group_23_strong 30 3719.4 3812.3 42.088 15.7913 4
Pr(>Chisq)
lcs_selfes_group_23_config
lcs_selfes_group_23_weak 0.284188
lcs_selfes_group_23_strong 0.003312 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_selfes_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_lcs_selfes_group_23_weak <- broom::tidy(lcs_selfes_group_23_weak, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_selfes_1 ~1 ")) %>% mutate(sig_diff = "**")
kable(params_lcs_selfes_group_23_weak, digits = 3)| term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_selfes_1 ~1 | d_int_a | 0.164 | 0.399 | 4.161 | 0.000 | ** |
| d_selfes_1 ~1 | d_int_b | 0.009 | 0.028 | 0.283 | 0.777 | ** |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(lcs_selfes_group_23_strong, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_selfes_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_selfes_1 ~1 | d_int | 0.07 | 0.166 | 2.517 | 0.012 |
| d_selfes_1 ~1 | d_int | 0.07 | 0.204 | 2.517 | 0.012 |
# whole model (weak invariance)
summary(lcs_selfes_group_23_weak, fit.measures=TRUE, standardized=TRUE, rsquare=F)lavaan 0.6.15 ended normally after 69 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 44
Number of equality constraints 16
Number of observations per group:
Group 2 179
Group 3 176
Number of missing patterns per group:
Group 2 2
Group 3 2
Model Test User Model:
Standard Scaled
Test Statistic 25.907 24.547
Degrees of freedom 26 26
P-value (Chi-square) 0.468 0.545
Scaling correction factor 1.055
Yuan-Bentler correction (Mplus variant)
Test statistic for each group:
Group 2 16.497 15.631
Group 3 9.409 8.915
Model Test Baseline Model:
Test statistic 1951.407 1562.615
Degrees of freedom 30 30
P-value 0.000 0.000
Scaling correction factor 1.249
User Model versus Baseline Model:
Comparative Fit Index (CFI) 1.000 1.000
Tucker-Lewis Index (TLI) 1.000 1.001
Robust Comparative Fit Index (CFI) 1.000
Robust Tucker-Lewis Index (TLI) 1.001
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -1827.611 -1827.611
Scaling correction factor 0.686
for the MLR correction
Loglikelihood unrestricted model (H1) -1814.658 -1814.658
Scaling correction factor 1.067
for the MLR correction
Akaike (AIC) 3711.222 3711.222
Bayesian (BIC) 3819.641 3819.641
Sample-size adjusted Bayesian (SABIC) 3730.813 3730.813
Root Mean Square Error of Approximation:
RMSEA 0.000 0.000
90 Percent confidence interval - lower 0.000 0.000
90 Percent confidence interval - upper 0.059 0.054
P-value H_0: RMSEA <= 0.050 0.890 0.926
P-value H_0: RMSEA >= 0.080 0.003 0.001
Robust RMSEA 0.000
90 Percent confidence interval - lower 0.000
90 Percent confidence interval - upper 0.060
P-value H_0: Robust RMSEA <= 0.050 0.888
P-value H_0: Robust RMSEA >= 0.080 0.005
Standardized Root Mean Square Residual:
SRMR 0.033 0.033
Parameter Estimates:
Standard errors Sandwich
Information bread Observed
Observed information based on Hessian
Group 1 [Group 2]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t1 =~
slf_1_1 1.000 0.726 0.874
slf_2_1 (lmb2) 1.164 0.037 31.311 0.000 0.846 0.911
slf_3_1 (lmb3) 1.145 0.048 23.709 0.000 0.832 0.795
selfes_t2 =~
slf_1_2 1.000 0.741 0.878
slf_2_2 (lmb2) 1.164 0.037 31.311 0.000 0.862 0.914
slf_3_2 (lmb3) 1.145 0.048 23.709 0.000 0.849 0.800
d_selfes_1 =~
slfs_t2 1.000 0.554 0.554
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t2 ~
selfes_t1 1.000 0.980 0.980
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t1 ~~
d_slf_1 (fb_a) -0.074 0.033 -2.248 0.025 -0.247 -0.247
.selfes_par1_t1 ~~
.slf_1_2 (cov1) 0.061 0.024 2.542 0.011 0.061 0.375
.selfes_par2_t1 ~~
.slf_2_2 (cov2) 0.059 0.028 2.129 0.033 0.059 0.402
.selfes_par3_t1 ~~
.slf_3_2 (cov3) 0.104 0.040 2.579 0.010 0.104 0.257
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.slfs_t2 0.000 0.000 0.000
d_slf_1 (d_n_) 0.164 0.039 4.161 0.000 0.399 0.399
slfs_t1 (wb__) 3.421 0.062 55.486 0.000 4.709 4.709
.slf_1_1 0.000 0.000 0.000
.slf_2_1 (m2a) -0.805 0.140 -5.738 0.000 -0.805 -0.867
.slf_3_1 (m3a) -1.184 0.183 -6.466 0.000 -1.184 -1.130
.slf_1_2 0.000 0.000 0.000
.slf_2_2 (m2a) -0.805 0.140 -5.738 0.000 -0.805 -0.853
.slf_3_2 (m3a) -1.184 0.183 -6.466 0.000 -1.184 -1.116
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.slfs_t2 0.000 0.000 0.000
d_slf_1 (d_v_) 0.169 0.035 4.826 0.000 1.000 1.000
slfs_t1 (wb__) 0.528 0.060 8.822 0.000 1.000 1.000
.slf_1_1 (res1) 0.163 0.023 7.103 0.000 0.163 0.236
.slf_2_1 (res2) 0.147 0.026 5.753 0.000 0.147 0.171
.slf_3_1 (res3) 0.404 0.041 9.790 0.000 0.404 0.368
.slf_1_2 (res1) 0.163 0.023 7.103 0.000 0.163 0.229
.slf_2_2 (res2) 0.147 0.026 5.753 0.000 0.147 0.165
.slf_3_2 (res3) 0.404 0.041 9.790 0.000 0.404 0.359
Group 2 [Group 3]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t1 =~
slf_1_1 1.000 0.812 0.899
slf_2_1 (lmb2) 1.164 0.037 31.311 0.000 0.945 0.945
slf_3_1 (lmb3) 1.145 0.048 23.709 0.000 0.930 0.846
selfes_t2 =~
slf_1_2 1.000 0.863 0.909
slf_2_2 (lmb2) 1.164 0.037 31.311 0.000 1.005 0.951
slf_3_2 (lmb3) 1.145 0.048 23.709 0.000 0.989 0.860
d_selfes_1 =~
slfs_t2 1.000 0.390 0.390
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t2 ~
selfes_t1 1.000 0.940 0.940
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
selfes_t1 ~~
d_slf_1 (fb_b) -0.014 0.029 -0.480 0.631 -0.050 -0.050
.selfes_par1_t1 ~~
.slf_1_2 (cv1b) 0.057 0.020 2.857 0.004 0.057 0.365
.selfes_par2_t1 ~~
.slf_2_2 (cv2b) 0.025 0.020 1.256 0.209 0.025 0.232
.selfes_par3_t1 ~~
.slf_3_2 (cv3b) 0.159 0.039 4.102 0.000 0.159 0.461
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.slfs_t2 0.000 0.000 0.000
d_slf_1 (d_n_) 0.009 0.033 0.283 0.777 0.028 0.028
slfs_t1 (wb__) 3.709 0.067 55.648 0.000 4.568 4.568
.slf_1_1 0.000 0.000 0.000
.slf_2_1 (m2b) -0.824 0.147 -5.598 0.000 -0.824 -0.824
.slf_3_1 (m3b) -1.207 0.188 -6.419 0.000 -1.207 -1.098
.slf_1_2 0.000 0.000 0.000
.slf_2_2 (m2b) -0.824 0.147 -5.598 0.000 -0.824 -0.780
.slf_3_2 (m3b) -1.207 0.188 -6.419 0.000 -1.207 -1.050
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.slfs_t2 0.000 0.000 0.000
d_slf_1 (d_v_) 0.114 0.029 3.872 0.000 1.000 1.000
slfs_t1 (wb__) 0.659 0.066 10.042 0.000 1.000 1.000
.slf_1_1 (rs1b) 0.157 0.021 7.430 0.000 0.157 0.192
.slf_2_1 (rs2b) 0.108 0.022 4.867 0.000 0.108 0.108
.slf_3_1 (rs3b) 0.345 0.038 9.122 0.000 0.345 0.285
.slf_1_2 (rs1b) 0.157 0.021 7.430 0.000 0.157 0.174
.slf_2_2 (rs2b) 0.108 0.022 4.867 0.000 0.108 0.096
.slf_3_2 (rs3b) 0.345 0.038 9.122 0.000 0.345 0.261Self-concept clarity: fitting multi-group models
# adapt latent change score model from above and add grouping factor in estimation (also add vectorized equality constraints to the model)
# configural invariance -> model template created above
lcs_concept_group_23_config <- sem(mi_lcs_concept_group_config, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr',
fixed.x=FALSE, missing='fiml', group = "group")
# weak invariance -> model template created above
lcs_concept_group_23_weak <- sem(mi_lcs_concept_group_weak, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
# strong invariance -> model template created above
lcs_concept_group_23_strong <- sem(mi_lcs_concept_group_strong, data=df_sbsa3_wide_wb %>%
filter(group!="Group 1"), estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))Self-concept clarity: results
# model comparison tests for measurement invariance
lavTestLRT(lcs_concept_group_23_config, lcs_concept_group_23_weak, lcs_concept_group_23_strong)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
lcs_concept_group_23_config 24 3897.8 4014.0 27.042
lcs_concept_group_23_weak 26 3896.1 4004.5 29.306 2.2311 2
lcs_concept_group_23_strong 30 3896.8 3989.7 38.030 8.5761 4
Pr(>Chisq)
lcs_concept_group_23_config
lcs_concept_group_23_weak 0.32773
lcs_concept_group_23_strong 0.07261 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_concept_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_lcs_concept_group_23_weak <- broom::tidy(lcs_concept_group_23_weak, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_concept_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_lcs_concept_group_23_weak, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_concept_1 ~1 | d_int_a | 0.076 | 0.179 | 1.888 | 0.059 | |
| d_concept_1 ~1 | d_int_b | 0.076 | 0.198 | 2.127 | 0.033 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(lcs_concept_group_23_strong, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_concept_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_concept_1 ~1 | d_int | 0.077 | 0.181 | 2.855 | 0.004 |
| d_concept_1 ~1 | d_int | 0.077 | 0.201 | 2.855 | 0.004 |
# whole model (weak invariance)
summary(lcs_concept_group_23_weak, fit.measures=TRUE, standardized=TRUE, rsquare=F)lavaan 0.6.15 ended normally after 62 iterations
Estimator ML
Optimization method NLMINB
Number of model parameters 44
Number of equality constraints 16
Number of observations per group:
Group 2 179
Group 3 176
Number of missing patterns per group:
Group 2 2
Group 3 2
Model Test User Model:
Standard Scaled
Test Statistic 29.306 28.349
Degrees of freedom 26 26
P-value (Chi-square) 0.297 0.341
Scaling correction factor 1.034
Yuan-Bentler correction (Mplus variant)
Test statistic for each group:
Group 2 9.046 8.751
Group 3 20.260 19.599
Model Test Baseline Model:
Test statistic 1705.074 1407.968
Degrees of freedom 30 30
P-value 0.000 0.000
Scaling correction factor 1.211
User Model versus Baseline Model:
Comparative Fit Index (CFI) 0.998 0.998
Tucker-Lewis Index (TLI) 0.998 0.998
Robust Comparative Fit Index (CFI) 0.998
Robust Tucker-Lewis Index (TLI) 0.998
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -1920.043 -1920.043
Scaling correction factor 0.671
for the MLR correction
Loglikelihood unrestricted model (H1) -1905.390 -1905.390
Scaling correction factor 1.045
for the MLR correction
Akaike (AIC) 3896.086 3896.086
Bayesian (BIC) 4004.505 4004.505
Sample-size adjusted Bayesian (SABIC) 3915.677 3915.677
Root Mean Square Error of Approximation:
RMSEA 0.027 0.023
90 Percent confidence interval - lower 0.000 0.000
90 Percent confidence interval - upper 0.067 0.064
P-value H_0: RMSEA <= 0.050 0.791 0.827
P-value H_0: RMSEA >= 0.080 0.010 0.007
Robust RMSEA 0.025
90 Percent confidence interval - lower 0.000
90 Percent confidence interval - upper 0.071
P-value H_0: Robust RMSEA <= 0.050 0.771
P-value H_0: Robust RMSEA >= 0.080 0.018
Standardized Root Mean Square Residual:
SRMR 0.039 0.039
Parameter Estimates:
Standard errors Sandwich
Information bread Observed
Observed information based on Hessian
Group 1 [Group 2]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t1 =~
cnc_1_1 1.000 0.668 0.794
cnc_2_1 (lmb2) 1.257 0.053 23.678 0.000 0.839 0.852
cnc_3_1 (lmb3) 1.290 0.055 23.296 0.000 0.861 0.913
concept_t2 =~
cnc_1_2 1.000 0.679 0.799
cnc_2_2 (lmb2) 1.257 0.053 23.678 0.000 0.853 0.855
cnc_3_2 (lmb3) 1.290 0.055 23.296 0.000 0.876 0.916
d_concept_1 =~
cncpt_2 1.000 0.624 0.624
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t2 ~
concept_t1 1.000 0.983 0.983
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t1 ~~
d_cnc_1 (fb_a) -0.082 0.027 -3.039 0.002 -0.291 -0.291
.concept_par1_t1 ~~
.cnc_1_2 (cov1) 0.077 0.027 2.850 0.004 0.077 0.296
.concept_par2_t1 ~~
.cnc_2_2 (cov2) 0.083 0.031 2.688 0.007 0.083 0.311
.concept_par3_t1 ~~
.cnc_3_2 (cov3) 0.049 0.029 1.650 0.099 0.049 0.328
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.cncpt_2 0.000 0.000 0.000
d_cnc_1 (d_n_) 0.076 0.040 1.888 0.059 0.179 0.179
cncpt_1 (wb__) 2.922 0.059 49.484 0.000 4.377 4.377
.cnc_1_1 0.000 0.000 0.000
.cnc_2_1 (m2a) -0.658 0.163 -4.033 0.000 -0.658 -0.668
.cnc_3_1 (m3a) -0.660 0.171 -3.866 0.000 -0.660 -0.699
.cnc_1_2 0.000 0.000 0.000
.cnc_2_2 (m2a) -0.658 0.163 -4.033 0.000 -0.658 -0.660
.cnc_3_2 (m3a) -0.660 0.171 -3.866 0.000 -0.660 -0.690
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.cncpt_2 0.000 0.000 0.000
d_cnc_1 (d_v_) 0.179 0.031 5.762 0.000 1.000 1.000
cncpt_1 (wb__) 0.446 0.054 8.327 0.000 1.000 1.000
.cnc_1_1 (res1) 0.261 0.025 10.624 0.000 0.261 0.369
.cnc_2_1 (res2) 0.267 0.035 7.540 0.000 0.267 0.275
.cnc_3_1 (res3) 0.148 0.030 4.933 0.000 0.148 0.166
.cnc_1_2 (res1) 0.261 0.025 10.624 0.000 0.261 0.362
.cnc_2_2 (res2) 0.267 0.035 7.540 0.000 0.267 0.268
.cnc_3_2 (res3) 0.148 0.030 4.933 0.000 0.148 0.162
Group 2 [Group 3]:
Latent Variables:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t1 =~
cnc_1_1 1.000 0.756 0.839
cnc_2_1 (lmb2) 1.257 0.053 23.678 0.000 0.950 0.896
cnc_3_1 (lmb3) 1.290 0.055 23.296 0.000 0.975 0.912
concept_t2 =~
cnc_1_2 1.000 0.754 0.838
cnc_2_2 (lmb2) 1.257 0.053 23.678 0.000 0.947 0.895
cnc_3_2 (lmb3) 1.290 0.055 23.296 0.000 0.972 0.911
d_concept_1 =~
cncpt_2 1.000 0.508 0.508
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t2 ~
concept_t1 1.000 1.003 1.003
Covariances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
concept_t1 ~~
d_cnc_1 (fb_b) -0.075 0.029 -2.539 0.011 -0.258 -0.258
.concept_par1_t1 ~~
.cnc_1_2 (cv1b) 0.078 0.025 3.073 0.002 0.078 0.322
.concept_par2_t1 ~~
.cnc_2_2 (cv2b) 0.052 0.035 1.506 0.132 0.052 0.235
.concept_par3_t1 ~~
.cnc_3_2 (cv3b) 0.103 0.035 2.978 0.003 0.103 0.535
Intercepts:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.cncpt_2 0.000 0.000 0.000
d_cnc_1 (d_n_) 0.076 0.036 2.127 0.033 0.198 0.198
cncpt_1 (wb__) 3.130 0.067 46.847 0.000 4.142 4.142
.cnc_1_1 0.000 0.000 0.000
.cnc_2_1 (m2b) -0.724 0.174 -4.170 0.000 -0.724 -0.683
.cnc_3_1 (m3b) -0.814 0.185 -4.391 0.000 -0.814 -0.761
.cnc_1_2 0.000 0.000 0.000
.cnc_2_2 (m2b) -0.724 0.174 -4.170 0.000 -0.724 -0.684
.cnc_3_2 (m3b) -0.814 0.185 -4.391 0.000 -0.814 -0.763
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.cncpt_2 0.000 0.000 0.000
d_cnc_1 (d_v_) 0.146 0.029 5.009 0.000 1.000 1.000
cncpt_1 (wb__) 0.571 0.064 8.981 0.000 1.000 1.000
.cnc_1_1 (rs1b) 0.241 0.027 8.778 0.000 0.241 0.296
.cnc_2_1 (rs2b) 0.223 0.031 7.223 0.000 0.223 0.198
.cnc_3_1 (rs3b) 0.193 0.034 5.721 0.000 0.193 0.169
.cnc_1_2 (rs1b) 0.241 0.027 8.778 0.000 0.241 0.297
.cnc_2_2 (rs2b) 0.223 0.031 7.223 0.000 0.223 0.199
.cnc_3_2 (rs3b) 0.193 0.034 5.721 0.000 0.193 0.170Adapt latent change score model from above (but without any moderation) and add grouping factor in estimation (adding vectorized equality constraints to the model step by step):
# create templates: (same as those already created above)
# loop across 5 traits
for (i in 1:5) {
item_nrs = b5_vars[[i]][[1]]
short_name = str_trunc(names(b5_vars)[i], 5, ellipsis = "")
# configural - current
template_filled_config_current <- str_replace_all(trait_template_main_config,
c("trait" = short_name,
"ind01" = paste0(short_name, "_curr_par1"),
"ind02" = paste0(short_name, "_curr_par2"),
"ind03" = paste0(short_name, "_curr_par3")))
trait_main_fit_config_current <- lavaan(template_filled_config_current,
data = df_sbsa3_wide_pers %>%
filter(group!="Group 1"),
estimator='mlr', fixed.x=FALSE, missing='fiml', group = "group")
eval(call("<-", as.name(paste0("lcs_", short_name, "_main_config_group_23")), template_filled_config_current))
eval(call("<-", as.name(paste0("fit_lcs_", short_name, "_main_config_group_23")), trait_main_fit_config_current))
# weak - current
template_filled_weak_current <- str_replace_all(trait_template_main_weak,
c("trait" = short_name,
"ind01" = paste0(short_name, "_curr_par1"),
"ind02" = paste0(short_name, "_curr_par2"),
"ind03" = paste0(short_name, "_curr_par3")))
trait_main_fit_weak_current <- lavaan(template_filled_weak_current,
data = df_sbsa3_wide_pers %>%
filter(group!="Group 1"),
estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = "loadings")
eval(call("<-", as.name(paste0("lcs_", short_name, "_main_weak_group_23")), template_filled_weak_current))
eval(call("<-", as.name(paste0("fit_lcs_", short_name, "_main_weak_group_23")), trait_main_fit_weak_current))
# strong - current
template_filled_strong_current <- str_replace_all(trait_template_main_strong,
c("trait" = short_name,
"ind01" = paste0(short_name, "_curr_par1"),
"ind02" = paste0(short_name, "_curr_par2"),
"ind03" = paste0(short_name, "_curr_par3")))
trait_main_fit_strong_current <- lavaan(template_filled_strong_current,
data = df_sbsa3_wide_pers %>%
filter(group!="Group 1"),
estimator='mlr', fixed.x=FALSE, missing='fiml',
group = "group", group.equal = c("intercepts", "loadings"))
eval(call("<-", as.name(paste0("lcs_", short_name, "_main_strong_group_23")), template_filled_strong_current))
eval(call("<-", as.name(paste0("fit_lcs_", short_name, "_main_strong_group_23")), trait_main_fit_strong_current))
}Extraversion (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_extra_main_config_group_23, fit_lcs_extra_main_weak_group_23, fit_lcs_extra_main_strong_group_23)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_extra_main_config_group_23 24 3313.7 3429.9 35.195
fit_lcs_extra_main_weak_group_23 26 3312.4 3420.9 37.936 2.9066 2
fit_lcs_extra_main_strong_group_23 30 3317.8 3410.7 51.292 12.9389 4
Pr(>Chisq)
fit_lcs_extra_main_config_group_23
fit_lcs_extra_main_weak_group_23 0.23380
fit_lcs_extra_main_strong_group_23 0.01158 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_extra_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_main_lcs_extra_group_23 <- broom::tidy(fit_lcs_extra_main_weak_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_extra_1 ~1 ")) %>% mutate(sig_diff = "*")
kable(params_main_lcs_extra_group_23, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_extra_1 ~1 | d_int_a | 0.091 | 0.284 | 2.694 | 0.007 | * |
| d_extra_1 ~1 | d_int_b | -0.039 | -0.258 | -1.530 | 0.126 | * |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_extra_main_strong_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_extra_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_extra_1 ~1 | d_int | 0.004 | 0.012 | 0.19 | 0.849 |
| d_extra_1 ~1 | d_int | 0.004 | 0.027 | 0.19 | 0.849 |
# model fit
kable(broom::glance(fit_lcs_extra_main_weak_group_23) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 355 | 28 | 3312.443 | 3420.862 | 0.993 | 0.051 | 0.045 |
Agreeableness (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_agree_main_config_group_23, fit_lcs_agree_main_weak_group_23, fit_lcs_agree_main_strong_group_23)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_agree_main_config_group_23 24 3226.1 3342.2 33.599
fit_lcs_agree_main_weak_group_23 26 3224.0 3332.4 35.570 2.0013 2
fit_lcs_agree_main_strong_group_23 30 3219.1 3312.1 38.679 2.9883 4
Pr(>Chisq)
fit_lcs_agree_main_config_group_23
fit_lcs_agree_main_weak_group_23 0.3676
fit_lcs_agree_main_strong_group_23 0.5598# show model with varying latent change parameters
# -> key parameter is "d_agree_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_main_lcs_agree_group_23 <- broom::tidy(fit_lcs_agree_main_weak_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_agree_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_main_lcs_agree_group_23, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_agree_1 ~1 | d_int_a | 0.044 | 0.170 | 1.603 | 0.109 | |
| d_agree_1 ~1 | d_int_b | -0.007 | -0.049 | -0.324 | 0.746 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_agree_main_strong_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_agree_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_agree_1 ~1 | d_int | 0.013 | 0.050 | 0.716 | 0.474 |
| d_agree_1 ~1 | d_int | 0.013 | 0.087 | 0.716 | 0.474 |
# model fit
kable(broom::glance(fit_lcs_agree_main_weak_group_23) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 355 | 28 | 3224.022 | 3332.442 | 0.992 | 0.046 | 0.04 |
Conscientiousness (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_consc_main_config_group_23, fit_lcs_consc_main_weak_group_23, fit_lcs_consc_main_strong_group_23)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_consc_main_config_group_23 24 3093.7 3209.9 24.840
fit_lcs_consc_main_weak_group_23 26 3090.4 3198.8 25.499 0.7707 2
fit_lcs_consc_main_strong_group_23 30 3095.5 3188.4 38.635 12.7522 4
Pr(>Chisq)
fit_lcs_consc_main_config_group_23
fit_lcs_consc_main_weak_group_23 0.68022
fit_lcs_consc_main_strong_group_23 0.01255 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_consc_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_main_lcs_consc_group_23 <- broom::tidy(fit_lcs_consc_main_weak_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_consc_1 ~1 ")) %>% mutate(sig_diff = "*")
kable(params_main_lcs_consc_group_23, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_consc_1 ~1 | d_int_a | 0.068 | 0.278 | 2.389 | 0.017 | * |
| d_consc_1 ~1 | d_int_b | 0.040 | 0.167 | 1.487 | 0.137 | * |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_consc_main_strong_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_consc_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_consc_1 ~1 | d_int | 0.052 | 0.211 | 2.617 | 0.009 |
| d_consc_1 ~1 | d_int | 0.052 | 0.219 | 2.617 | 0.009 |
# model fit
kable(broom::glance(fit_lcs_consc_main_weak_group_23) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 355 | 28 | 3090.36 | 3198.779 | 1 | 0 | 0.029 |
Neuroticism (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_neuro_main_config_group_23, fit_lcs_neuro_main_weak_group_23, fit_lcs_neuro_main_strong_group_23)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_neuro_main_config_group_23 24 3165.7 3281.8 38.800
fit_lcs_neuro_main_weak_group_23 26 3162.0 3270.5 39.162 0.3657 2
fit_lcs_neuro_main_strong_group_23 30 3170.2 3263.1 55.319 16.8305 4
Pr(>Chisq)
fit_lcs_neuro_main_config_group_23
fit_lcs_neuro_main_weak_group_23 0.832882
fit_lcs_neuro_main_strong_group_23 0.002085 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# show model with varying latent change parameters
# -> key parameter is "d_neuro_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_main_lcs_neuro_group_23 <- broom::tidy(fit_lcs_neuro_main_weak_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_neuro_1 ~1 ")) %>% mutate(sig_diff = "**")
kable(params_main_lcs_neuro_group_23, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_neuro_1 ~1 | d_int_a | -0.125 | -0.459 | -4.268 | 0.000 | ** |
| d_neuro_1 ~1 | d_int_b | 0.018 | 0.059 | 0.578 | 0.563 | ** |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_neuro_main_strong_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_neuro_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_neuro_1 ~1 | d_int | -0.055 | -0.194 | -2.344 | 0.019 |
| d_neuro_1 ~1 | d_int | -0.055 | -0.179 | -2.344 | 0.019 |
# model fit
kable(broom::glance(fit_lcs_neuro_main_weak_group_23) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 355 | 28 | 3162.035 | 3270.454 | 0.994 | 0.053 | 0.025 |
Openness (current personality): results
# model comparison tests for measurement invariance
lavTestLRT(fit_lcs_openn_main_config_group_23, fit_lcs_openn_main_weak_group_23, fit_lcs_openn_main_strong_group_23)
Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
lavaan NOTE:
The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference
test is a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff
fit_lcs_openn_main_config_group_23 24 3290.0 3406.1 22.233
fit_lcs_openn_main_weak_group_23 26 3287.8 3396.3 24.097 1.9657 2
fit_lcs_openn_main_strong_group_23 30 3284.3 3377.2 28.526 4.2923 4
Pr(>Chisq)
fit_lcs_openn_main_config_group_23
fit_lcs_openn_main_weak_group_23 0.3742
fit_lcs_openn_main_strong_group_23 0.3679# show model with varying latent change parameters
# -> key parameter is "d_openn_1 ~1"
# labelled parameter as "d_int_a" & "d_int_b" (a = Group 2, b = Group 3)
params_main_lcs_openn_group_23 <- broom::tidy(fit_lcs_openn_main_weak_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_openn_1 ~1 ")) %>% mutate(sig_diff = " ")
kable(params_main_lcs_openn_group_23, digits = 3) | term | label | estimate | std.all | statistic | p.value | sig_diff |
|---|---|---|---|---|---|---|
| d_openn_1 ~1 | d_int_a | 0.042 | 0.217 | 1.766 | 0.077 | |
| d_openn_1 ~1 | d_int_b | -0.001 | -0.013 | -0.035 | 0.972 |
# constrained to be equal in the strong measurement invariance model:
kable(broom::tidy(fit_lcs_openn_main_strong_group_23, conf.int = TRUE, conf.level = 0.99) %>%
select(term, label, estimate, std.all, statistic, p.value) %>%
filter(term %in% c("d_openn_1 ~1 ")), digits = 3) | term | label | estimate | std.all | statistic | p.value |
|---|---|---|---|---|---|
| d_openn_1 ~1 | d_int | 0.016 | 0.084 | 1.096 | 0.273 |
| d_openn_1 ~1 | d_int | 0.016 | 0.313 | 1.096 | 0.273 |
# model fit
kable(broom::glance(fit_lcs_openn_main_weak_group_23) %>%
select(nobs, npar, AIC, BIC, cfi, rmsea, srmr), digits = 3)| nobs | npar | AIC | BIC | cfi | rmsea | srmr |
|---|---|---|---|---|---|---|
| 355 | 28 | 3287.837 | 3396.256 | 1 | 0 | 0.04 |
params_main_summary_12 <- bind_rows(
params_main_lcs_extra_group_12 %>% mutate(trait = names(b5_vars)[1], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff),
params_main_lcs_agree_group_12 %>% mutate(trait = names(b5_vars)[2], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff),
params_main_lcs_consc_group_12 %>% mutate(trait = names(b5_vars)[3], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff),
params_main_lcs_neuro_group_12 %>% mutate(trait = names(b5_vars)[4], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff),
params_main_lcs_openn_group_12 %>% mutate(trait = names(b5_vars)[5], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff)) %>% mutate(group = ifelse(group=="d_int_a", "Group 1", "Group 2"))
kable(params_main_summary_12, digits = 3)| trait | self | group | beta | p.value | sig_diff |
|---|---|---|---|---|---|
| extraversion | current | Group 1 | 0.370 | 0.006 | |
| extraversion | current | Group 2 | 0.290 | 0.006 | |
| agreeableness | current | Group 1 | 0.176 | 0.084 | |
| agreeableness | current | Group 2 | 0.168 | 0.105 | |
| conscientiousness | current | Group 1 | 0.326 | 0.003 | |
| conscientiousness | current | Group 2 | 0.276 | 0.018 | |
| neuroticism | current | Group 1 | -0.145 | 0.151 | |
| neuroticism | current | Group 2 | -0.462 | 0.000 | |
| openness | current | Group 1 | 0.110 | 0.412 | |
| openness | current | Group 2 | 0.221 | 0.072 |
params_main_summary_23 <- bind_rows(
params_main_lcs_extra_group_23 %>% mutate(trait = names(b5_vars)[1], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff),
params_main_lcs_agree_group_23 %>% mutate(trait = names(b5_vars)[2], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff),
params_main_lcs_consc_group_23 %>% mutate(trait = names(b5_vars)[3], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff),
params_main_lcs_neuro_group_23 %>% mutate(trait = names(b5_vars)[4], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff),
params_main_lcs_openn_group_23 %>% mutate(trait = names(b5_vars)[5], self = "current") %>% select(trait, self, group = label, beta = std.all, p.value, sig_diff)) %>% mutate(group = ifelse(group=="d_int_a", "Group 2", "Group 3"))
kable(params_main_summary_23, digits = 3)| trait | self | group | beta | p.value | sig_diff |
|---|---|---|---|---|---|
| extraversion | current | Group 2 | 0.284 | 0.007 | * |
| extraversion | current | Group 3 | -0.258 | 0.126 | * |
| agreeableness | current | Group 2 | 0.170 | 0.109 | |
| agreeableness | current | Group 3 | -0.049 | 0.746 | |
| conscientiousness | current | Group 2 | 0.278 | 0.017 | * |
| conscientiousness | current | Group 3 | 0.167 | 0.137 | * |
| neuroticism | current | Group 2 | -0.459 | 0.000 | ** |
| neuroticism | current | Group 3 | 0.059 | 0.563 | ** |
| openness | current | Group 2 | 0.217 | 0.077 | |
| openness | current | Group 3 | -0.013 | 0.972 |
params_main_summary_wb_12 <- bind_rows(
params_lcs_swls_group_12_weak %>% mutate(trait = "Life Satisfaction") %>% select(trait, group = label, beta = std.all, p.value, sig_diff),
params_lcs_meaning_group_12_weak %>% mutate(trait = "Meaning in Life") %>% select(trait, group = label, beta = std.all, p.value, sig_diff),
params_lcs_selfes_group_12_weak %>% mutate(trait = "Self-Esteem") %>% select(trait, group = label, beta = std.all, p.value, sig_diff),
params_lcs_concept_group_12_weak %>% mutate(trait = "Self-Concept Clarity") %>% select(trait, group = label, beta = std.all, p.value, sig_diff)) %>% mutate(group = ifelse(group=="d_int_a", "Group 1", "Group 2"))
kable(params_main_summary_wb_12, digits = 3)| trait | group | beta | p.value | sig_diff |
|---|---|---|---|---|
| Life Satisfaction | Group 1 | 0.317 | 0.001 | |
| Life Satisfaction | Group 2 | 0.512 | 0.000 | |
| Meaning in Life | Group 1 | 0.186 | 0.050 | |
| Meaning in Life | Group 2 | 0.200 | 0.042 | |
| Self-Esteem | Group 1 | 0.247 | 0.006 | ** |
| Self-Esteem | Group 2 | 0.398 | 0.000 | ** |
| Self-Concept Clarity | Group 1 | 0.324 | 0.000 | |
| Self-Concept Clarity | Group 2 | 0.181 | 0.057 |
params_main_summary_wb_23 <- bind_rows(
params_lcs_swls_group_23_weak %>% mutate(trait = "Life Satisfaction") %>% select(trait, group = label, beta = std.all, p.value, sig_diff),
params_lcs_meaning_group_23_weak %>% mutate(trait = "Meaning in Life") %>% select(trait, group = label, beta = std.all, p.value, sig_diff),
params_lcs_selfes_group_23_weak %>% mutate(trait = "Self-Esteem") %>% select(trait, group = label, beta = std.all, p.value, sig_diff),
params_lcs_concept_group_23_weak %>% mutate(trait = "Self-Concept Clarity") %>% select(trait, group = label, beta = std.all, p.value, sig_diff)) %>% mutate(group = ifelse(group=="d_int_a", "Group 2", "Group 3"))
kable(params_main_summary_wb_23, digits = 3)| trait | group | beta | p.value | sig_diff |
|---|---|---|---|---|
| Life Satisfaction | Group 2 | 0.511 | 0.000 | ** |
| Life Satisfaction | Group 3 | 0.034 | 0.720 | ** |
| Meaning in Life | Group 2 | 0.196 | 0.046 | |
| Meaning in Life | Group 3 | 0.146 | 0.115 | |
| Self-Esteem | Group 2 | 0.399 | 0.000 | ** |
| Self-Esteem | Group 3 | 0.028 | 0.777 | ** |
| Self-Concept Clarity | Group 2 | 0.179 | 0.059 | |
| Self-Concept Clarity | Group 3 | 0.198 | 0.033 |
Gather all latent intercepts at T1 and latent change scores:
b5_latent_df_plot <- bind_rows(
broom::tidy(fit_lcs_extra_main_weak_group_12, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("extra_t1 ~1 ", "d_extra_1 ~1 ")) %>%
mutate(Trait = "Extraversion", Group = ifelse(label %in% c("d_int_a", "b5_int_a"), "Group 1", "Group 2")),
broom::tidy(fit_lcs_extra_main_weak_group_23, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("extra_t1 ~1 ", "d_extra_1 ~1 ") & label %in% c("d_int_b", "b5_int_b")) %>%
mutate(Trait = "Extraversion", Group = "Group 3"),
broom::tidy(fit_lcs_agree_main_weak_group_12, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("agree_t1 ~1 ", "d_agree_1 ~1 ")) %>%
mutate(Trait = "Agreeableness", Group = ifelse(label %in% c("d_int_a", "b5_int_a"), "Group 1", "Group 2")),
broom::tidy(fit_lcs_agree_main_weak_group_23, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("agree_t1 ~1 ", "d_agree_1 ~1 ") & label %in% c("d_int_b", "b5_int_b")) %>%
mutate(Trait = "Agreeableness", Group = "Group 3"),
broom::tidy(fit_lcs_consc_main_weak_group_12, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("consc_t1 ~1 ", "d_consc_1 ~1 ")) %>%
mutate(Trait = "Conscientiousness", Group = ifelse(label %in% c("d_int_a", "b5_int_a"), "Group 1", "Group 2")),
broom::tidy(fit_lcs_consc_main_weak_group_23, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("consc_t1 ~1 ", "d_consc_1 ~1 ") & label %in% c("d_int_b", "b5_int_b")) %>%
mutate(Trait = "Conscientiousness", Group = "Group 3"),
broom::tidy(fit_lcs_neuro_main_weak_group_12, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("neuro_t1 ~1 ", "d_neuro_1 ~1 ")) %>%
mutate(Trait = "Neuroticism", Group = ifelse(label %in% c("d_int_a", "b5_int_a"), "Group 1", "Group 2")),
broom::tidy(fit_lcs_neuro_main_weak_group_23, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("neuro_t1 ~1 ", "d_neuro_1 ~1 ") & label %in% c("d_int_b", "b5_int_b")) %>%
mutate(Trait = "Neuroticism", Group = "Group 3"),
broom::tidy(fit_lcs_openn_main_weak_group_12, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("openn_t1 ~1 ", "d_openn_1 ~1 ")) %>%
mutate(Trait = "Openness", Group = ifelse(label %in% c("d_int_a", "b5_int_a"), "Group 1", "Group 2")),
broom::tidy(fit_lcs_openn_main_weak_group_23, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("openn_t1 ~1 ", "d_openn_1 ~1 ") & label %in% c("d_int_b", "b5_int_b")) %>%
mutate(Trait = "Openness", Group = "Group 3"))
# select and mutate relevant vars
b5_latent_df_plot <- b5_latent_df_plot %>%
mutate(param = ifelse(label %in% c("d_int_a", "d_int_b"), "slope", "intercept")) %>%
select(Trait, Group, estimate, std.error, param, conf.low, conf.high) %>%
pivot_wider(names_from = param,
names_sep = "_",
values_from = -c(Trait, Group, param)) %>%
mutate(T2_estimate = estimate_intercept + estimate_slope, T1_estimate = estimate_intercept,
T2_cilow = T2_estimate - qnorm(0.995)*(sqrt(std.error_intercept^2 + std.error_slope^2)),
T2_cihi = T2_estimate + qnorm(0.995)*(sqrt(std.error_intercept^2 + std.error_slope^2))) %>%
rename(T1_cilow = conf.low_intercept, T1_cihi = conf.high_intercept) %>%
select(-c(estimate_slope, estimate_intercept, conf.low_slope, conf.high_slope, std.error_slope, std.error_intercept))
# reshape
b5_latent_df_plot <- b5_latent_df_plot %>%
pivot_longer(-c(Trait, Group),
names_to = c("Time", ".value"),
names_pattern = "T(\\d)_(.*)") %>%
mutate(Time = factor(str_c("T", Time), levels = c("T1", "T2"), labels = c("T1", "T2"), ordered = T))Plot changes (unstandardized effect size metric): all together
ggplot(b5_latent_df_plot,
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2.5, 4.5) +
facet_wrap( ~ Trait, ncol = 3) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Extraversion
ggplot(b5_latent_df_plot %>% filter(Trait=="Extraversion"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2.5, 3.5) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Agreeableness
ggplot(b5_latent_df_plot %>% filter(Trait=="Agreeableness"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(3.5, 4.5) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Conscientiousness
ggplot(b5_latent_df_plot %>% filter(Trait=="Conscientiousness"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2.75, 3.75) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Neuroticism
ggplot(b5_latent_df_plot %>% filter(Trait=="Neuroticism"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2.5, 3.5) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Openness
ggplot(b5_latent_df_plot %>% filter(Trait=="Openness"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(3, 4) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Gather all latent intercepts at T1 and latent change scores:
wb_latent_df_plot <- bind_rows(
broom::tidy(lcs_swls_group_12_weak, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("swls_t1 ~1 ", "d_swls_1 ~1 ")) %>%
mutate(Trait = "Life Satisfaction", Group = ifelse(label %in% c("d_int_a", "wb_int_a"), "Group 1", "Group 2")),
broom::tidy(lcs_swls_group_23_weak, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("swls_t1 ~1 ", "d_swls_1 ~1 ") & label %in% c("d_int_b", "wb_int_b")) %>%
mutate(Trait = "Life Satisfaction", Group = "Group 3"),
broom::tidy(lcs_meaning_group_12_weak, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("meaning_t1 ~1 ", "d_meaning_1 ~1 ")) %>%
mutate(Trait = "Meaning in Life", Group = ifelse(label %in% c("d_int_a", "wb_int_a"), "Group 1", "Group 2")),
broom::tidy(lcs_meaning_group_23_weak, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("meaning_t1 ~1 ", "d_meaning_1 ~1 ") & label %in% c("d_int_b", "wb_int_b")) %>%
mutate(Trait = "Meaning in Life", Group = "Group 3"),
broom::tidy(lcs_selfes_group_12_weak, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("selfes_t1 ~1 ", "d_selfes_1 ~1 ")) %>%
mutate(Trait = "Self-Esteem", Group = ifelse(label %in% c("d_int_a", "wb_int_a"), "Group 1", "Group 2")),
broom::tidy(lcs_selfes_group_23_weak, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("selfes_t1 ~1 ", "d_selfes_1 ~1 ") & label %in% c("d_int_b", "wb_int_b")) %>%
mutate(Trait = "Self-Esteem", Group = "Group 3"),
broom::tidy(lcs_concept_group_12_weak, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("concept_t1 ~1 ", "d_concept_1 ~1 ")) %>%
mutate(Trait = "Self-Concept Clarity", Group = ifelse(label %in% c("d_int_a", "wb_int_a"), "Group 1", "Group 2")),
broom::tidy(lcs_concept_group_23_weak, conf.int = T, conf.level = .99) %>% select(term, label, estimate, std.error, conf.low, conf.high) %>%
filter(term %in% c("concept_t1 ~1 ", "d_concept_1 ~1 ") & label %in% c("d_int_b", "wb_int_b")) %>%
mutate(Trait = "Self-Concept Clarity", Group = "Group 3"))
# select and mutate relevant vars
wb_latent_df_plot <- wb_latent_df_plot %>%
mutate(param = ifelse(label %in% c("d_int_a", "d_int_b"), "slope", "intercept")) %>%
select(Trait, Group, estimate, std.error, param, conf.low, conf.high) %>%
pivot_wider(names_from = param,
names_sep = "_",
values_from = -c(Trait, Group, param)) %>%
mutate(T2_estimate = estimate_intercept + estimate_slope, T1_estimate = estimate_intercept,
T2_cilow = T2_estimate - qnorm(0.995)*(sqrt(std.error_intercept^2 + std.error_slope^2)),
T2_cihi = T2_estimate + qnorm(0.995)*(sqrt(std.error_intercept^2 + std.error_slope^2))) %>%
rename(T1_cilow = conf.low_intercept, T1_cihi = conf.high_intercept) %>%
select(-c(estimate_slope, estimate_intercept, conf.low_slope, conf.high_slope, std.error_slope, std.error_intercept))
# reshape
wb_latent_df_plot <- wb_latent_df_plot %>%
pivot_longer(-c(Trait, Group),
names_to = c("Time", ".value"),
names_pattern = "T(\\d)_(.*)") %>%
mutate(Time = factor(str_c("T", Time), levels = c("T1", "T2"), labels = c("T1", "T2"), ordered = T))Plot changes (unstandardized effect size metric): all together
ggplot(wb_latent_df_plot,
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2, 4) +
facet_wrap( ~ Trait, ncol = 2) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Life Satisfaction
ggplot(wb_latent_df_plot %>% filter(Trait=="Life Satisfaction"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2, 3.5) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Meaning in Life
ggplot(wb_latent_df_plot %>% filter(Trait=="Meaning in Life"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2.5, 4) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Self-Esteem
ggplot(wb_latent_df_plot %>% filter(Trait=="Self-Esteem"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2.5, 4) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Plotting model based mean-level changes: Self-Concept Clarity
ggplot(wb_latent_df_plot %>% filter(Trait=="Self-Concept Clarity"),
aes(y = estimate, x = Time)) +
geom_point(aes(shape = Group, group = Group, color = Group), position = position_dodge(0.3)) +
geom_line(aes(group = Group, color = Group, linetype = Group), position = position_dodge(0.3)) +
scale_shape_manual(values=c(17, 19, 15)) +
geom_errorbar(aes(group = Group, color = Group, ymin = cilow, ymax = cihi), width=.3, position=position_dodge(0.3)) +
ylab("Mean Trait Level\n(Model-Based, 99% CI)") +
ylim(2.5, 4) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
strip.background=element_rect(fill="grey93"))
Do changes in current personality trait change depend on the study group? Using multilevel change models instead of SEM latent change score models.
Changes in current personality traits over time:
df_manip_check_curr_12 <- df_sbsa3 %>%
filter(group!="Group 3") %>%
mutate(time_d = time - 1) %>%
select(pid, time_d, group,
paste0(str_trunc(names(b5_vars), 5, ellipsis = ""), "_comb_curr")) %>%
pivot_longer(ends_with("_comb_curr"),
names_to = "test", names_prefix = "facet", values_to = "score", values_drop_na = TRUE) %>%
mutate(group_1 = ifelse(group=="Group 1", 1, 0)) %>%
group_nest(test) %>%
mutate(lmer_mods = map(data, ~lmerTest::lmer(score ~ time_d * group_1 + (1 | pid), data = .x))) %>%
pull(lmer_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_manip_check_curr_unlist_12 <- as.data.frame(summary(df_manip_check_curr_12[[1]])$coefficients) %>% as_tibble()
for (i in 2:length(df_manip_check_curr_12)) {
df_manip_check_curr_unlist_12 <- bind_rows(df_manip_check_curr_unlist_12,
as.data.frame(summary(df_manip_check_curr_12[[i]])$coefficients) %>% as_tibble())
}kable(df_manip_check_curr_unlist_12 %>%
mutate(outcome = rep(sort(names(b5_vars)), each=4),
term = c(rep(c("Intercept", "time", "group_1", "time*group_1"), 20))) %>%
rename(p = `Pr(>|t|)`, str_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, str_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars)))), # order by BFI traits,
digits = 3)| outcome | term | est | str_err | p | sig |
|---|---|---|---|---|---|
| extraversion | Intercept | 2.921 | 0.056 | 0.000 | *** |
| extraversion | time | 0.090 | 0.031 | 0.003 | ** |
| extraversion | group_1 | -0.089 | 0.080 | 0.266 | |
| extraversion | time*group_1 | -0.014 | 0.043 | 0.743 | |
| agreeableness | Intercept | 3.615 | 0.043 | 0.000 | *** |
| agreeableness | time | 0.052 | 0.034 | 0.125 | |
| agreeableness | group_1 | 0.072 | 0.061 | 0.237 | |
| agreeableness | time*group_1 | -0.002 | 0.048 | 0.961 | |
| conscientiousness | Intercept | 3.364 | 0.055 | 0.000 | *** |
| conscientiousness | time | 0.063 | 0.032 | 0.048 | * |
| conscientiousness | group_1 | -0.034 | 0.078 | 0.657 | |
| conscientiousness | time*group_1 | 0.036 | 0.045 | 0.424 | |
| neuroticism | Intercept | 3.263 | 0.063 | 0.000 | *** |
| neuroticism | time | -0.143 | 0.035 | 0.000 | *** |
| neuroticism | group_1 | -0.074 | 0.090 | 0.412 | |
| neuroticism | time*group_1 | 0.087 | 0.049 | 0.078 | |
| openness | Intercept | 3.704 | 0.048 | 0.000 | *** |
| openness | time | 0.055 | 0.029 | 0.061 | |
| openness | group_1 | 0.027 | 0.068 | 0.687 | |
| openness | time*group_1 | -0.028 | 0.041 | 0.497 | |
| sociability | Intercept | 2.634 | 0.075 | 0.000 | *** |
| sociability | time | 0.140 | 0.041 | 0.001 | *** |
| sociability | group_1 | -0.179 | 0.106 | 0.093 | |
| sociability | time*group_1 | -0.008 | 0.058 | 0.886 | |
| assertiveness | Intercept | 2.966 | 0.067 | 0.000 | *** |
| assertiveness | time | 0.025 | 0.045 | 0.584 | |
| assertiveness | group_1 | -0.041 | 0.095 | 0.663 | |
| assertiveness | time*group_1 | -0.008 | 0.064 | 0.903 | |
| energy | Intercept | 3.162 | 0.064 | 0.000 | *** |
| energy | time | 0.106 | 0.046 | 0.022 | * |
| energy | group_1 | -0.046 | 0.091 | 0.611 | |
| energy | time*group_1 | -0.029 | 0.065 | 0.650 | |
| compassion | Intercept | 3.779 | 0.053 | 0.000 | *** |
| compassion | time | -0.014 | 0.051 | 0.779 | |
| compassion | group_1 | 0.020 | 0.075 | 0.792 | |
| compassion | time*group_1 | 0.134 | 0.072 | 0.064 | |
| respectfulness | Intercept | 3.955 | 0.051 | 0.000 | *** |
| respectfulness | time | 0.062 | 0.043 | 0.145 | |
| respectfulness | group_1 | 0.099 | 0.072 | 0.171 | |
| respectfulness | time*group_1 | -0.050 | 0.060 | 0.401 | |
| trust | Intercept | 3.112 | 0.060 | 0.000 | *** |
| trust | time | 0.114 | 0.048 | 0.020 | * |
| trust | group_1 | 0.099 | 0.085 | 0.243 | |
| trust | time*group_1 | -0.090 | 0.068 | 0.190 | |
| organization | Intercept | 3.446 | 0.075 | 0.000 | *** |
| organization | time | 0.045 | 0.047 | 0.340 | |
| organization | group_1 | -0.043 | 0.107 | 0.689 | |
| organization | time*group_1 | 0.036 | 0.067 | 0.590 | |
| productiveness | Intercept | 3.161 | 0.065 | 0.000 | *** |
| productiveness | time | 0.103 | 0.048 | 0.033 | * |
| productiveness | group_1 | -0.132 | 0.093 | 0.156 | |
| productiveness | time*group_1 | 0.038 | 0.068 | 0.580 | |
| responsibility | Intercept | 3.486 | 0.057 | 0.000 | *** |
| responsibility | time | 0.049 | 0.044 | 0.265 | |
| responsibility | group_1 | 0.072 | 0.081 | 0.379 | |
| responsibility | time*group_1 | 0.025 | 0.062 | 0.693 | |
| anxiety | Intercept | 3.602 | 0.069 | 0.000 | *** |
| anxiety | time | -0.090 | 0.046 | 0.049 | * |
| anxiety | group_1 | -0.069 | 0.098 | 0.484 | |
| anxiety | time*group_1 | 0.065 | 0.064 | 0.316 | |
| depression | Intercept | 3.176 | 0.073 | 0.000 | *** |
| depression | time | -0.216 | 0.046 | 0.000 | *** |
| depression | group_1 | -0.086 | 0.104 | 0.413 | |
| depression | time*group_1 | 0.141 | 0.065 | 0.032 | * |
| volatility | Intercept | 3.011 | 0.074 | 0.000 | *** |
| volatility | time | -0.127 | 0.053 | 0.017 | * |
| volatility | group_1 | -0.067 | 0.106 | 0.527 | |
| volatility | time*group_1 | 0.060 | 0.074 | 0.422 | |
| curiosity | Intercept | 3.880 | 0.054 | 0.000 | *** |
| curiosity | time | 0.075 | 0.040 | 0.059 | |
| curiosity | group_1 | -0.008 | 0.077 | 0.914 | |
| curiosity | time*group_1 | -0.030 | 0.056 | 0.597 | |
| aesthetic | Intercept | 3.599 | 0.062 | 0.000 | *** |
| aesthetic | time | 0.025 | 0.046 | 0.589 | |
| aesthetic | group_1 | -0.008 | 0.089 | 0.930 | |
| aesthetic | time*group_1 | -0.008 | 0.065 | 0.898 | |
| imagination | Intercept | 3.633 | 0.060 | 0.000 | *** |
| imagination | time | 0.069 | 0.046 | 0.137 | |
| imagination | group_1 | 0.100 | 0.086 | 0.246 | |
| imagination | time*group_1 | -0.048 | 0.066 | 0.464 |
Plot of main effects: extraversion and neuroticism
ggeffects::ggpredict(df_manip_check_curr_12$extraversion,
terms = c("time_d[0,1]", "group_1[0,1]")) |> plot() +
aes(linetype=group, color=group, shape=group) + geom_point() +
theme_light() +
scale_x_continuous(breaks = c(0,1), labels = c("T1", "T2")) +
theme(panel.grid.minor.x = element_blank()) +
labs(title = "Extraversion", x = "Time", y = "Predicted values of\nextraversion",
linetype="Group\n(0 = Group 2,\n1 = Group 1)", color="Group\n(0 = Group 2,\n1 = Group 1)",
shape="Group\n(0 = Group 2,\n1 = Group 1)")
ggeffects::ggpredict(df_manip_check_curr_12$neuroticism,
terms = c("time_d[0,1]", "group_1[0,1]")) |> plot() +
aes(linetype=group, color=group, shape=group) + geom_point() +
theme_light() +
scale_x_continuous(breaks = c(0,1), labels = c("T1", "T2")) +
theme(panel.grid.minor.x = element_blank()) +
labs(title = "Neuroticism", x = "Time", y = "Predicted values of\nneuroticism",
linetype="Group\n(0 = Group 2,\n1 = Group 1)", color="Group\n(0 = Group 2,\n1 = Group 1)",
shape="Group\n(0 = Group 2,\n1 = Group 1)")
Group interaction effect: Depression (but only at p < .05)
ggeffects::ggpredict(df_manip_check_curr_12$depression,
terms = c("time_d[0,1]", "group_1[0,1]")) |> plot() +
aes(linetype=group, color=group, shape=group) + geom_point() +
theme_light() +
scale_x_continuous(breaks = c(0,1), labels = c("T1", "T2")) +
theme(panel.grid.minor.x = element_blank()) +
labs(title = "Depression", x = "Time", y = "Predicted values of\ndepression",
linetype="Group\n(0 = Group 2,\n1 = Group 1)", color="Group\n(0 = Group 2,\n1 = Group 1)",
shape="Group\n(0 = Group 2,\n1 = Group 1)")
Changes in ideal personality traits over time:
df_manip_check_ideal <- df_sbsa3 %>%
filter(group=="Group 1") %>%
mutate(time_d = time - 1) %>%
select(pid, time_d,
paste0(str_trunc(names(b5_vars), 5, ellipsis = ""), "_comb_ideal")) %>%
pivot_longer(ends_with("_comb_ideal"),
names_to = "test", names_prefix = "facet", values_to = "score", values_drop_na = TRUE) %>%
group_nest(test) %>%
mutate(lmer_mods = map(data, ~lmerTest::lmer(score ~ time_d + (1 | pid), data = .x))) %>%
pull(lmer_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_manip_check_ideal_unlist <- as.data.frame(summary(df_manip_check_ideal[[1]])$coefficients) %>% as_tibble()
for (i in 2:length(df_manip_check_ideal)) {
df_manip_check_ideal_unlist <- bind_rows(df_manip_check_ideal_unlist,
as.data.frame(summary(df_manip_check_ideal[[i]])$coefficients) %>% as_tibble())
}kable(df_manip_check_ideal_unlist %>%
mutate(outcome = rep(sort(names(b5_vars)), each=2),
term = c(rep(c("Intercept", "time"), 20))) %>%
rename(p = `Pr(>|t|)`, str_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, str_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars)))), # order by BFI traits,
digits = 3)| outcome | term | est | str_err | p | sig |
|---|---|---|---|---|---|
| extraversion | Intercept | 3.917 | 0.039 | 0.000 | *** |
| extraversion | time | -0.102 | 0.036 | 0.005 | ** |
| agreeableness | Intercept | 4.069 | 0.040 | 0.000 | *** |
| agreeableness | time | -0.023 | 0.033 | 0.473 | |
| conscientiousness | Intercept | 4.391 | 0.034 | 0.000 | *** |
| conscientiousness | time | 0.005 | 0.033 | 0.886 | |
| neuroticism | Intercept | 1.802 | 0.036 | 0.000 | *** |
| neuroticism | time | 0.031 | 0.034 | 0.370 | |
| openness | Intercept | 4.115 | 0.042 | 0.000 | *** |
| openness | time | 0.015 | 0.035 | 0.675 | |
| sociability | Intercept | 3.647 | 0.056 | 0.000 | *** |
| sociability | time | -0.125 | 0.053 | 0.019 | * |
| assertiveness | Intercept | 3.889 | 0.052 | 0.000 | *** |
| assertiveness | time | -0.152 | 0.047 | 0.002 | ** |
| energy | Intercept | 4.217 | 0.044 | 0.000 | *** |
| energy | time | -0.031 | 0.049 | 0.521 | |
| compassion | Intercept | 4.017 | 0.052 | 0.000 | *** |
| compassion | time | 0.021 | 0.055 | 0.708 | |
| respectfulness | Intercept | 4.462 | 0.043 | 0.000 | *** |
| respectfulness | time | -0.053 | 0.042 | 0.209 | |
| trust | Intercept | 3.728 | 0.055 | 0.000 | *** |
| trust | time | -0.037 | 0.049 | 0.457 | |
| organization | Intercept | 4.414 | 0.043 | 0.000 | *** |
| organization | time | 0.008 | 0.048 | 0.874 | |
| productiveness | Intercept | 4.506 | 0.041 | 0.000 | *** |
| productiveness | time | -0.013 | 0.047 | 0.775 | |
| responsibility | Intercept | 4.254 | 0.045 | 0.000 | *** |
| responsibility | time | 0.024 | 0.044 | 0.585 | |
| anxiety | Intercept | 1.902 | 0.047 | 0.000 | *** |
| anxiety | time | 0.080 | 0.054 | 0.144 | |
| depression | Intercept | 1.511 | 0.041 | 0.000 | *** |
| depression | time | 0.033 | 0.041 | 0.425 | |
| volatility | Intercept | 1.991 | 0.045 | 0.000 | *** |
| volatility | time | -0.024 | 0.048 | 0.614 | |
| curiosity | Intercept | 4.151 | 0.050 | 0.000 | *** |
| curiosity | time | -0.033 | 0.050 | 0.509 | |
| aesthetic | Intercept | 3.863 | 0.053 | 0.000 | *** |
| aesthetic | time | 0.078 | 0.050 | 0.126 | |
| imagination | Intercept | 4.334 | 0.053 | 0.000 | *** |
| imagination | time | 0.002 | 0.056 | 0.968 |
Changes in current personality traits over time:
df_manip_check_curr_23 <- df_sbsa3 %>%
filter(group!="Group 1") %>%
mutate(time_d = time - 1) %>%
select(pid, time_d, group,
paste0(str_trunc(names(b5_vars), 5, ellipsis = ""), "_comb_curr")) %>%
pivot_longer(ends_with("_comb_curr"),
names_to = "test", names_prefix = "facet", values_to = "score", values_drop_na = TRUE) %>%
mutate(group_2 = ifelse(group=="Group 2", 1, 0)) %>%
group_nest(test) %>%
mutate(lmer_mods = map(data, ~lmerTest::lmer(score ~ time_d * group_2 + (1 | pid), data = .x))) %>%
pull(lmer_mods) %>%
purrr::set_names(sort(names(b5_vars))) # ordered alphabetically
df_manip_check_curr_unlist_23 <- as.data.frame(summary(df_manip_check_curr_23[[1]])$coefficients) %>% as_tibble()
for (i in 2:length(df_manip_check_curr_23)) {
df_manip_check_curr_unlist_23 <- bind_rows(df_manip_check_curr_unlist_23,
as.data.frame(summary(df_manip_check_curr_23[[i]])$coefficients) %>% as_tibble())
}kable(df_manip_check_curr_unlist_23 %>%
mutate(outcome = rep(sort(names(b5_vars)), each=4),
term = c(rep(c("Intercept", "time", "group_2", "time*group_2"), 20))) %>%
rename(p = `Pr(>|t|)`, str_err = `Std. Error`, est = Estimate) %>% select(outcome, term, est, str_err, p) %>%
mutate(sig = ifelse(p < .05, ifelse(p < .01, ifelse(p < .001, "***", "**"), "*"), " ")) %>%
arrange(factor(outcome, levels = (names(b5_vars)))), # order by BFI traits,
digits = 3)| outcome | term | est | str_err | p | sig |
|---|---|---|---|---|---|
| extraversion | Intercept | 3.069 | 0.058 | 0.000 | *** |
| extraversion | time | -0.036 | 0.029 | 0.207 | |
| extraversion | group_2 | -0.148 | 0.081 | 0.070 | |
| extraversion | time*group_2 | 0.127 | 0.041 | 0.002 | ** |
| agreeableness | Intercept | 3.689 | 0.046 | 0.000 | *** |
| agreeableness | time | -0.005 | 0.030 | 0.876 | |
| agreeableness | group_2 | -0.073 | 0.065 | 0.256 | |
| agreeableness | time*group_2 | 0.056 | 0.043 | 0.201 | |
| conscientiousness | Intercept | 3.543 | 0.056 | 0.000 | *** |
| conscientiousness | time | 0.042 | 0.028 | 0.139 | |
| conscientiousness | group_2 | -0.179 | 0.078 | 0.023 | * |
| conscientiousness | time*group_2 | 0.021 | 0.040 | 0.605 | |
| neuroticism | Intercept | 3.032 | 0.067 | 0.000 | *** |
| neuroticism | time | 0.014 | 0.031 | 0.648 | |
| neuroticism | group_2 | 0.231 | 0.095 | 0.015 | * |
| neuroticism | time*group_2 | -0.157 | 0.045 | 0.001 | *** |
| openness | Intercept | 3.834 | 0.049 | 0.000 | *** |
| openness | time | -0.008 | 0.024 | 0.749 | |
| openness | group_2 | -0.130 | 0.070 | 0.063 | |
| openness | time*group_2 | 0.063 | 0.035 | 0.075 | |
| sociability | Intercept | 2.696 | 0.078 | 0.000 | *** |
| sociability | time | 0.014 | 0.039 | 0.720 | |
| sociability | group_2 | -0.062 | 0.109 | 0.571 | |
| sociability | time*group_2 | 0.126 | 0.056 | 0.025 | * |
| assertiveness | Intercept | 3.173 | 0.068 | 0.000 | *** |
| assertiveness | time | -0.061 | 0.041 | 0.144 | |
| assertiveness | group_2 | -0.207 | 0.096 | 0.032 | * |
| assertiveness | time*group_2 | 0.086 | 0.059 | 0.148 | |
| energy | Intercept | 3.337 | 0.067 | 0.000 | *** |
| energy | time | -0.062 | 0.045 | 0.167 | |
| energy | group_2 | -0.175 | 0.095 | 0.065 | |
| energy | time*group_2 | 0.167 | 0.064 | 0.009 | ** |
| compassion | Intercept | 3.807 | 0.057 | 0.000 | *** |
| compassion | time | -0.018 | 0.048 | 0.710 | |
| compassion | group_2 | -0.027 | 0.081 | 0.733 | |
| compassion | time*group_2 | 0.001 | 0.068 | 0.989 | |
| respectfulness | Intercept | 4.009 | 0.051 | 0.000 | *** |
| respectfulness | time | 0.036 | 0.040 | 0.365 | |
| respectfulness | group_2 | -0.053 | 0.072 | 0.461 | |
| respectfulness | time*group_2 | 0.026 | 0.057 | 0.652 | |
| trust | Intercept | 3.251 | 0.062 | 0.000 | *** |
| trust | time | -0.032 | 0.042 | 0.457 | |
| trust | group_2 | -0.140 | 0.088 | 0.112 | |
| trust | time*group_2 | 0.143 | 0.061 | 0.019 | * |
| organization | Intercept | 3.685 | 0.074 | 0.000 | *** |
| organization | time | -0.049 | 0.041 | 0.238 | |
| organization | group_2 | -0.239 | 0.104 | 0.022 | * |
| organization | time*group_2 | 0.093 | 0.059 | 0.114 | |
| productiveness | Intercept | 3.345 | 0.066 | 0.000 | *** |
| productiveness | time | 0.139 | 0.042 | 0.001 | ** |
| productiveness | group_2 | -0.185 | 0.092 | 0.047 | * |
| productiveness | time*group_2 | -0.038 | 0.060 | 0.529 | |
| responsibility | Intercept | 3.598 | 0.056 | 0.000 | *** |
| responsibility | time | 0.033 | 0.041 | 0.421 | |
| responsibility | group_2 | -0.112 | 0.079 | 0.157 | |
| responsibility | time*group_2 | 0.016 | 0.058 | 0.785 | |
| anxiety | Intercept | 3.412 | 0.073 | 0.000 | *** |
| anxiety | time | 0.035 | 0.042 | 0.405 | |
| anxiety | group_2 | 0.190 | 0.102 | 0.064 | |
| anxiety | time*group_2 | -0.124 | 0.060 | 0.038 | * |
| depression | Intercept | 2.827 | 0.077 | 0.000 | *** |
| depression | time | 0.022 | 0.043 | 0.612 | |
| depression | group_2 | 0.349 | 0.108 | 0.001 | ** |
| depression | time*group_2 | -0.237 | 0.062 | 0.000 | *** |
| volatility | Intercept | 2.857 | 0.077 | 0.000 | *** |
| volatility | time | -0.016 | 0.046 | 0.735 | |
| volatility | group_2 | 0.155 | 0.109 | 0.157 | |
| volatility | time*group_2 | -0.110 | 0.066 | 0.097 | |
| curiosity | Intercept | 4.006 | 0.055 | 0.000 | *** |
| curiosity | time | -0.001 | 0.034 | 0.973 | |
| curiosity | group_2 | -0.125 | 0.077 | 0.106 | |
| curiosity | time*group_2 | 0.075 | 0.048 | 0.120 | |
| aesthetic | Intercept | 3.722 | 0.065 | 0.000 | *** |
| aesthetic | time | -0.010 | 0.038 | 0.782 | |
| aesthetic | group_2 | -0.122 | 0.091 | 0.179 | |
| aesthetic | time*group_2 | 0.035 | 0.054 | 0.520 | |
| imagination | Intercept | 3.774 | 0.063 | 0.000 | *** |
| imagination | time | -0.008 | 0.043 | 0.860 | |
| imagination | group_2 | -0.141 | 0.089 | 0.112 | |
| imagination | time*group_2 | 0.076 | 0.061 | 0.216 |
Plot of main effects: productiveness
ggeffects::ggpredict(df_manip_check_curr_23$productiveness,
terms = c("time_d[0,1]", "group_2[0,1]")) |> plot() +
aes(linetype=group, color=group, shape=group) + geom_point() +
theme_light() +
scale_x_continuous(breaks = c(0,1), labels = c("T1", "T2")) +
theme(panel.grid.minor.x = element_blank()) +
labs(title = "Productiveness", x = "Time", y = "Predicted values of\nproductiveness",
linetype="Group\n(0 = Group 3,\n1 = Group 2)", color="Group\n(0 = Group 3,\n1 = Group 2)",
shape="Group\n(0 = Group 3,\n1 = Group 2)")
Group interaction effects: extraversion, neuroticism, and depression
ggeffects::ggpredict(df_manip_check_curr_23$extraversion,
terms = c("time_d[0,1]", "group_2[0,1]")) |> plot() +
aes(linetype=group, color=group, shape=group) + geom_point() +
theme_light() +
scale_x_continuous(breaks = c(0,1), labels = c("T1", "T2")) +
theme(panel.grid.minor.x = element_blank()) +
labs(title = "Extraversion", x = "Time", y = "Predicted values of\nextraversion",
linetype="Group\n(0 = Group 3,\n1 = Group 2)", color="Group\n(0 = Group 3,\n1 = Group 2)",
shape="Group\n(0 = Group 3,\n1 = Group 2)")
ggeffects::ggpredict(df_manip_check_curr_23$neuroticism,
terms = c("time_d[0,1]", "group_2[0,1]")) |> plot() +
aes(linetype=group, color=group, shape=group) + geom_point() +
theme_light() +
scale_x_continuous(breaks = c(0,1), labels = c("T1", "T2")) +
theme(panel.grid.minor.x = element_blank()) +
labs(title = "Neuroticism", x = "Time", y = "Predicted values of\nneuroticism",
linetype="Group\n(0 = Group 3,\n1 = Group 2)", color="Group\n(0 = Group 3,\n1 = Group 2)",
shape="Group\n(0 = Group 3,\n1 = Group 2)")
ggeffects::ggpredict(df_manip_check_curr_23$depression,
terms = c("time_d[0,1]", "group_2[0,1]")) |> plot() +
aes(linetype=group, color=group, shape=group) + geom_point() +
theme_light() +
scale_x_continuous(breaks = c(0,1), labels = c("T1", "T2")) +
theme(panel.grid.minor.x = element_blank()) +
labs(title = "Depression", x = "Time", y = "Predicted values of\ndepression",
linetype="Group\n(0 = Group 3,\n1 = Group 2)", color="Group\n(0 = Group 3,\n1 = Group 2)",
shape="Group\n(0 = Group 3,\n1 = Group 2)")
df_sbsa3 <- df_sbsa3 %>% mutate(interv = ifelse(group=="Group 3", 0, 1))
# extra
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(extra_comb_curr),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 3.07 0.76 3.08 3.05 0.8 1 5 4 0.15 -0.46 0.06
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 2.88 0.75 2.83 2.86 0.74 1 5 4 0.18 -0.26 0.04t.test(extra_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: extra_comb_curr by interv
t = 2.7368, df = 347.5, p-value = 0.006522
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
0.01026637 0.37255887
sample estimates:
mean in group 0 mean in group 1
3.068655 2.877243 cohen.d(extra_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = extra_comb_curr ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
extra_comb_curr -0.49 -0.25 -0.01
Multivariate (Mahalanobis) distance between groups
[1] 0.25
r equivalent of difference between two means
extra_comb_curr
-0.12 # agree
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(agree_comb_curr),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 3.69 0.62 3.71 3.72 0.56 1.75 5 3.25 -0.35 0.06 0.05
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 3.65 0.57 3.67 3.67 0.62 1.75 4.92 3.17 -0.39 0.06 0.03t.test(agree_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: agree_comb_curr by interv
t = 0.64373, df = 326.78, p-value = 0.5202
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
-0.1087340 0.1806259
sample estimates:
mean in group 0 mean in group 1
3.688920 3.652975 cohen.d(agree_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = agree_comb_curr ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
agree_comb_curr -0.3 -0.06 0.18
Multivariate (Mahalanobis) distance between groups
[1] 0.061
r equivalent of difference between two means
agree_comb_curr
-0.03 # consc
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(consc_comb_curr),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 3.54 0.77 3.58 3.58 0.74 1.75 5 3.25 -0.41 -0.45 0.06
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 3.35 0.74 3.42 3.37 0.74 1.17 5 3.83 -0.26 -0.2 0.04t.test(consc_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: consc_comb_curr by interv
t = 2.7852, df = 338.3, p-value = 0.005651
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
0.0136436 0.3765884
sample estimates:
mean in group 0 mean in group 1
3.542614 3.347498 cohen.d(consc_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = consc_comb_curr ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
consc_comb_curr -0.5 -0.26 -0.02
Multivariate (Mahalanobis) distance between groups
[1] 0.26
r equivalent of difference between two means
consc_comb_curr
-0.12 # neuro
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(neuro_comb_curr),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 3.03 0.94 3 3.04 1.11 1.08 4.92 3.83 -0.03 -0.89 0.07
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 3.23 0.86 3.25 3.25 0.86 1.25 5 3.75 -0.23 -0.59 0.05t.test(neuro_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: neuro_comb_curr by interv
t = -2.3139, df = 322.55, p-value = 0.0213
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
-0.41266814 0.02332946
sample estimates:
mean in group 0 mean in group 1
3.031723 3.226393 cohen.d(neuro_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = neuro_comb_curr ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
neuro_comb_curr -0.02 0.22 0.46
Multivariate (Mahalanobis) distance between groups
[1] 0.22
r equivalent of difference between two means
neuro_comb_curr
0.1 # openn
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(openn_comb_curr),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 3.83 0.64 3.83 3.85 0.62 2.08 5 2.92 -0.23 -0.56 0.05
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 3.72 0.64 3.67 3.74 0.62 1.75 5 3.25 -0.33 -0.06 0.03t.test(openn_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: openn_comb_curr by interv
t = 1.9307, df = 350.41, p-value = 0.05433
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
-0.03895913 0.26713590
sample estimates:
mean in group 0 mean in group 1
3.833807 3.719718 cohen.d(openn_comb_curr ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = openn_comb_curr ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
openn_comb_curr -0.42 -0.18 0.06
Multivariate (Mahalanobis) distance between groups
[1] 0.18
r equivalent of difference between two means
openn_comb_curr
-0.08 # swls
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(swls),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 2.94 1 3 2.94 1.19 1 5 4 -0.06 -0.8 0.08
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 2.69 1 2.6 2.66 1.19 1 5 4 0.18 -0.83 0.05t.test(swls ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: swls by interv
t = 2.7266, df = 349.3, p-value = 0.006721
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
0.01260011 0.49016193
sample estimates:
mean in group 0 mean in group 1
2.937500 2.686119 cohen.d(swls ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = swls ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
swls -0.49 -0.25 -0.01
Multivariate (Mahalanobis) distance between groups
[1] 0.25
r equivalent of difference between two means
swls
-0.12 # meaning
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(meaning),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 3.77 1.17 3.8 3.77 1.04 1 7 6 -0.05 -0.01 0.09
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 3.67 1.12 3.6 3.61 1.04 1 7 6 0.52 0.24 0.06t.test(meaning ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: meaning by interv
t = 0.87265, df = 336.74, p-value = 0.3835
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
-0.1832919 0.3695107
sample estimates:
mean in group 0 mean in group 1
3.767614 3.674504 cohen.d(meaning ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = meaning ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
meaning -0.32 -0.08 0.16
Multivariate (Mahalanobis) distance between groups
[1] 0.082
r equivalent of difference between two means
meaning
-0.04 # selfes
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(selfes),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 3.44 0.93 3.5 3.46 1.04 1.2 5 3.8 -0.17 -0.95 0.07
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 3.23 0.86 3.1 3.23 0.89 1.1 5 3.9 0.07 -0.67 0.05t.test(selfes ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: selfes by interv
t = 2.5001, df = 323.94, p-value = 0.01291
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
-0.007620598 0.426558919
sample estimates:
mean in group 0 mean in group 1
3.442614 3.233144 cohen.d(selfes ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = selfes ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
selfes -0.48 -0.24 0
Multivariate (Mahalanobis) distance between groups
[1] 0.24
r equivalent of difference between two means
selfes
-0.11 # concept
psych::describeBy(df_sbsa3 %>% filter(time==1) %>% pull(concept),
group = df_sbsa3 %>% filter(time==1) %>% pull(interv))
Descriptive statistics by group
group: 0
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 176 3.16 0.91 3.17 3.16 1.05 1.25 5 3.75 0.05 -0.91 0.07
------------------------------------------------------------
group: 1
vars n mean sd median trimmed mad min max range skew kurtosis se
X1 1 353 3.06 0.85 3 3.03 0.86 1.33 5 3.67 0.32 -0.73 0.05t.test(concept ~ interv, data = df_sbsa3 %>% filter(time==1), paired = F, conf.level = 0.99)
Welch Two Sample t-test
data: concept by interv
t = 1.1862, df = 328.92, p-value = 0.2364
alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
99 percent confidence interval:
-0.1159972 0.3119046
sample estimates:
mean in group 0 mean in group 1
3.160985 3.063031 cohen.d(concept ~ interv, data = df_sbsa3 %>% filter(time==1), alpha=.01)Call: cohen.d(x = concept ~ interv, alpha = 0.01, data = df_sbsa3 %>%
filter(time == 1))
Cohen d statistic of difference between two means
lower effect upper
concept -0.35 -0.11 0.13
Multivariate (Mahalanobis) distance between groups
[1] 0.11
r equivalent of difference between two means
concept
-0.05